Find Your Dream Career

FUNCTION & DUTIES:Under the Assistant Director of Complementary Services, the Consultant will work in close collaboration with the other special education consultants in all areas of special education to provide services to special needs students in Inuktitut and to contribute to the in-service training and course sessions of Inuit teachers. Assessment and diagnosis of students with suspected learning problems in the schoolsIdentification of students with emotional and/or behavioural problemsTraining and support for the special education teachers and counsellors within the BoardOn-going development of culturally appropriate diagnostic tools and remedial materialIn-service and teacher training in the areas of educationInvolvement in on-going research concerning language of instruction. QUALIFICATIONS:Valid teaching diploma in the province of Quebec or close to graduating from the Kativik-McGill Teacher Training ProgramAt least three (3) years of teaching experience with children with special needs. The Board may, at its discretion, waive any or all of the afore-mentioned qualifications if it finds a suitable candidate who is a beneficiary of the James Bay and Northern Québec Agreement (JBNQA) and who accepts, as a condition of employment, to follow a training plan determined by the Board. REQUIREMENTS:Fluency in written and spoken Inuktitut and a good command of a second language (French or English)Knowledge of diagnostic and remedial tools in InuktitutAvailable to travel regularly. CLASSIFICATION / SALARY / BENEFITS:In accordance with the SPPENOM collective agreement for professionals. This position belongs to job category 2104: from $56,670 TO $105,432 annually depending on qualification and experience. All benefits are prorated based on the percentage of the task.

If you are committed to public service, enjoy collaborating with others, share our values and have a desire to learn and grow, join The City of Calgary. City employees deliver the services, run the programs and operate the facilities which make a difference in our community. We support work-life balance, promote physical and psychological safety, and offer competitive wages, pensions, and benefits. Together we make Calgary a great place to make a living, a great place to make a life. The City is committed to fostering a respectful, inclusive and equitable workplace which is representative of the community we serve. We welcome those who have demonstrated a commitment to upholding the values of equity, diversity, inclusion, anti-racism and reconciliation. Applications are encouraged from members of groups that are historically disadvantaged and underrepresented. Accommodations are available during the hiring process, upon request. As the Vehicle and Equipment Training Coordinator, you will be scheduling all vehicle and equipment training for Mobility operations, including recertifications, remedial training, equipment upgrades, and new staff onboarding. In this position, you will support multiple operational groups across 15 locations and 20 shifts and also support a variety of adult learning activities related to orientation, training rollouts, learning initiatives, and employee development. You will also maintain training records and supports compliance with training, CUPE 37 collective agreement, and MOU requirements. Primary duties include:Coordinate one-on-one vehicle and equipment training across Mobility operational groups by creating efficient training schedules, prioritizing requests, balancing operational readiness, trainer availability, vendor capacity, equipment constraints, employee release time, and shift requirements.Work with internal leaders, employees, Trainers, and vendors to clarify training needs, provide updates, resolve scheduling issues, manage expectations, and adjust training plans based on changing operational priorities.Input relevant course data into the LMS and use LMS information to create training gap reports that support leader decision-making.Track training activity by division, equipment type, employee qualification status, and training priority to support compliance and seasonal readiness goals. Develop and maintain training status tracking tools, dashboards, and summary reports that give leaders visibility into training progress, outstanding requests, queue status, employee training needs, and upcoming training priorities.Coordinate prerequisite, compliance-based, recertification, corrective, and seasonal training needs, including the rollout of updated training documentation and verification of employee training acknowledgements when operational, safety, or equipment-related issues are identified. Qualifications A completed 2 year Diploma in Business Administration, Adult Learning, Event Management, or a related field and at least 3 years of related experience.Experience in an organizational effectiveness or learning and development role is required.Intermediate level of proficiency in Microsoft Office (Word, PowerPoint and Excel) is also required.Proficiency working with PeopleSoft Human Capital Management (HCM), Sum Total Learning Management System, SharePoint and Livelink would be considered an asset.Success in this position requires the following:Customer service and communication skills.Ability to work with people at all levels of the organization, solve problems both independently and as part of a larger team.Strong organizational skills and attention to detail are essential. Pre-employment Requirements Successful applicants must provide proof of qualifications.Workstyle This position may be eligible to work from home for at least part of the time as one of several flexible work options available to City employees. These arrangements depend on the operational requirements of the role,

FUNCTION & DUTIES:Under the supervision of the Assistant Director General, the Centre Director is the senior administrator of the Board in his/her community and is responsible for all aspects of the organization and administration of the community school and its related services. Other duties include:Responsible for the implementation of Board and local policy as determined by the Commissioners and the local Education Committee.Is the liaison between the community and theResponsible for the organization and supervision of the physical plant including school, residences, shops, warehouse, vehicles and equipment.Is the Secretary-Treasurer of the local EducationSupervise, mentor and manage employees under their responsibility.Is responsible for all aspects of the organization and administration of Board activities within his/herQUALIFICATIONS:Undergraduate degree in a relevant field of study certifying a minimum three (3)-year university program or hold a senior or senior executive staff position, excluding a position as manager, in a board.The Board may, at its discretion, waive any or all of the aforementioned qualification requirements if it finds a suitable candidate who is a beneficiary of the James Bay and Northern Québec Agreement and who accepts, as a condition of employment, to follow a training plan determined by the Board. REQUIREMENTS:Fluency in Inuktitut and either French or EnglishA minimum of three (3) years of experience in the field of administration or education or equivalent relevant experience recognized by the BoardMust have been a full-time resident of an Inuit community of Nunavik for at least ten (10) yearsAn understanding of the Quebec Education system and Chapter 17 of the James Bay and Northern Quebec Agreement is an asset. CLASSIFICATION / SALARY / BENEFITS:Class 00-08 as per the working conditions of managers at KI: from $100,606 to $134,139 annually depending on qualifications and experience. All benefits are prorated based on the percentage of the task. No housing available for this position

Education: Earned doctorate degree. Engineering, general. Area of work experience: Teaching. Research. Personal suitability: Accurate. Dependability. Efficient interpersonal skills. Excellent oral communication. Excellent written communication. Flexibility. Initiative. Judgement. Organized. Reliability. Team player. Innovation. Screening questions: Are you available to start on the date listed in the job posting?. Do you have experience working in this field?. Do you have the required certifications listed in the job posting?. Do you meet the language requirements listed in the job posting for the position (English or French)?. Experience: Experience an asset. Workplace information: Hybrid. Health benefits: Disability benefits. Health care plan. Paramedical services coverage. Vision care benefits. Financial benefits: As per collective agreement. Group insurance benefits. Life insurance. Pension plan. Long term benefits: Long-term care insurance. Maternity and parental benefits. Other benefits: Learning/training paid by employer. On-site amenities. On-site daycare available. On-site recreation and activities. Other benefits. Paid time off (volunteering or personal days). Team building opportunities. Parking available. Travel insurance. Wellness program. Support for persons with disabilities: Provides physical accessibility accommodations (for example: ramps, elevators, etc.). Provides visual accessibility accommodations (for example: braille, screen readers, etc.). Provides auditory accessibility accommodations (for example: transcription software, teletypewriters, etc.). Offers mentorship, coaching and/or networking opportunities for persons with disabilities . Provides awareness training to employees to create a welcoming work environment for persons with disabilities. Applies accessible and inclusive recruitment policies that accommodate persons with disabilities . Support for Indigenous people: Offers mentorship, coaching and/or networking opportunities for Indigenous workers . Develops and maintains relationships with indigenous communities, indigenous-owned businesses and organizations . Provides cultural competency training and/or awareness training to all employees to create a welcoming work environment for Indigenous workers . Supports for visible minorities: Applies hiring policies that discourage discrimination against members of visible minorities (for example: anonymizing the hiring process, etc.). Offers mentorship programs that pair members of visible minorities with experienced employees. Provides diversity and cross-cultural training to create a welcoming work environment for members of visible minorities.

About Nebius: Nebius is leading a new era in cloud infrastructure for the global AI economy. We are building a full-stack AI cloud platform that supports developers and enterprises from data and model training through to production deployment, without the cost and complexity of building large in-house AI/ML infrastructure. Built by engineers, for engineers. From large-scale GPU orchestration to inference optimization, we own the hard problems across compute, storage, networking and applied AI. Listed on Nasdaq (NBIS) and headquartered in Amsterdam, we have a global footprint with R&D hubs across Europe, the UK, North America and Israel. Our team of 1,500+ includes hundreds of engineers with deep expertise across hardware, software and AI R&D. Role Summary Nebius Token Factory is building production-grade inference for open-source, customer, and internally optimized models. This role owns the path from model artifact to fast, reliable, cost-efficient endpoint. The work sits at the intersection of model internals, inference engines, serving architecture, benchmarking, and production deployment. What You Will Do Own a platform-level optimization pillar such as quantization, speculative decoding, KV -cache optimization, engine selection, or long-context serving. Create reusable evaluation, benchmarking, and rollout frameworks across multiple model families. Provide technical direction to MLEs, Applied Scientists, GPU kernel engineers, and infrastructure engineers on shared performance goals. Optimize LLM and VLM endpoints for latency, throughput, memory efficiency, GPU utilization, quality, and cost per token. Deploy, configure, benchmark, and extend inference engines such as vLLM, SGLang, TensorRT-LLM, Triton Inference Server, NVIDIA Dynamo, or similar systems. Build and productionize model-compression workflows, including quantization, quantization-aware training, distillation, low-bit serving, and accuracy recovery. Implement or integrate speculative decoding, draft-model approaches, KV -cache optimization, prefix caching, chunked prefill, continuous batching, and disaggregated prefill/decode serving. Build reproducible benchmark harnesses for TTFT , TPOT , tokens per second per GPU, p95/p99 latency, GPU memory, reliability, and cost per token. Partner with GPU kernel engineers and platform engineers to diagnose bottlenecks across model code, kernels, runtime, scheduler, gateway, and cluster layers. Write clear design docs, performance reports, rollout plans, and customer-facing technical explanations. Required Qualifications Strong Python and PyTorch engineering skills. Hands-on experience deploying or optimizing LLM, VLM , or high-throughput transformer inference systems. Practical knowledge of at least one modern inference stack such as vLLM, SGLang, TensorRT-LLM, Triton Inference Server, NVIDIA Dynamo, Ray Serve, KServe, or equivalent internal systems. Strong understanding of transformer inference bottlenecks, including KV cache, attention, memory bandwidth, batching, parallelism, and long-context serving. Ability to reason quantitatively about latency, throughput, quality, utilization, and cost tradeoffs. Strong communication skills and ability to collaborate with research, kernel, infrastructure, product, and customer teams. Preferred Qualifications Experience with quantization-aware training, post-training quantization, FP8 , INT8 , INT4 , NVFP4 , MXFP4 , AWQ , GPTQ , SmoothQuant, or related techniques. Experience with distillation, speculative decoding, EAGLE, Medusa, multi-token prediction, or other inference acceleration methods. Experience with agentic workloads, including tool calling, structured outputs, streaming APIs, high concurrency, and multi-step orchestration. CUDA or Triton familiarity, even if the role is not primarily a kernel-engineering role. Open-source contributions to vLLM, SGLang, TensorRT-LLM, FlashInfer, LMCache, PyTorch, Triton, Ray, KServe, or related projects. Benefits & Perks: Competitive compensation Career growth and learning opportunities Flexibility and ownership Collaborative and innovative culture Opportunity to work on impactful AI projects International environment and talented teams What's it like to work at Nebius: Fast moving - Bold thinking - Constant growth - Meaningful impact - Trust and real ownership - Opportunity to shape the future of AI Equal Opportunity Statement: Nebius is an equal opportunity employer. We are committed to fostering an inclusive and diverse workplace and to providing equal employment opportunities in all aspects of employment. We do not discriminate on the basis of race, color, religion, sex (including pregnancy), national origin, ancestry, age, disability, genetic information, marital status, veteran status, sexual orientation, gender identity or expression, or any other characteristic protected by applicable law. Applicants must be authorized to work in the country in which they apply and will be required to provide proof of employment eligibility as a condition of hire. If you need accommodations during the application process, please let us know. Find more English Speaking Jobs in United Kingdom on Arbeitnow

About us EggAI is a fast-growing AI consulting and development startup. We build agentic AI solutions for enterprise clients. We are growing in several European markets. The role You will lead our clients through the organisational and operational changes needed to establish a shared AI platform. The platform has two parts: an AI Control Plane that governs what agents may do and records the evidence, and a managed AI Runtime where the agents run. Both integrate with the client's identity, security, delivery and audit services, and both need owners, approval routes and support arrangements on the client side. You will own delivery outcomes from mobilisation through production release and operating handover, acting as the client sponsor's day-to-day counterpart. The role reaches across engineering, business leadership, security, risk, compliance and operations. You will bring these groups together to agree how the platform is governed and supported, and what evidence they need for assurance in a regulated environment. You will lead an EggAI delivery team with a mix of Platform Engineering roles and skills. Client application teams develop the business use cases; you lead AI platform delivery and the changes needed for teams to adopt and operate it. Why this role Most organisations are still working out how to let AI agents act safely in production: who approves what an agent may do, how its actions are evidenced, and who answers when something goes wrong. There is no settled playbook yet, especially in regulated industries. In this role you will help write it. The operating models, approval routes and assurance practices you design with our clients will shape how they run agentic AI for years, and they will become the basis of how EggAI delivers across Europe. You will work at the point where AI capability meets enterprise reality, with direct access to client sponsors and EggAI leadership, and with real influence over how a growing company builds its delivery practice. Responsibilities - Own the delivery plan and milestones, keeping commitments achievable within the agreed scope, budget and team capacity. - Secure the client commitments that delivery depends on. Agree who will provide access, complete service changes and make decisions, then resolve delays with the responsible teams. - Design the target operating model with the client. Define who owns the platform service, maintains policies and software, supports the runtime and approves changes, with agreed escalation routes. - Run milestone acceptance. Submit evidence for each milestone, secure recorded client decisions within agreed review periods and escalate overdue or disputed decisions. - Lead change control with the sponsor and EggAI leadership. Agree scope, capacity, cost and date changes, including proposals for additional services or extended support, before committing the team. - Secure production readiness. Confirm the service owners and funded operating cover from the operating model are in place before go-live, escalating any gap that prevents release. - Lead go-live coordination and the intensive support period after release. Bring client operations and application teams together to resolve incidents and agree when normal support can take over. - Lead platform adoption and training with the technical team, preparing client teams to use shared controls and follow the agreed delivery processes. - Run operating handover through practical exercises. Secure acceptance of ongoing duties and agree any further support needed before the current engagement ends. - Give sponsors a clear account of progress and delivery risks through weekly delivery reviews, regular demonstrations and sponsor reviews. Report spending, forecast work and adoption results so they can make timely decisions. What You Bring - You have led technology programmes into production in a regulated environment, including operational readiness and handover to receiving teams. - You have designed and introduced operating models, turning agreed responsibilities into working approval, support and service-management processes. - You have led delivery across engineering teams, client departments and external partners, including influencing senior sponsors and resolving competing priorities. - You have managed budgets and contractual scope in client delivery, including negotiating changes and shaping additional work. - You understand cloud platforms, security and service operations well enough to challenge plans and dependencies with technical specialists. - You communicate clearly and can lead workshops that turn disagreement or uncertainty into decisions and action. Nice to have Delivery in financial services or banking. AI platforms or agentic AI. Platform engineering, identity, security controls or audit and assurance work. Cloud platform delivery using Azure (preferred), AWS or Google Cloud, and AI platform services such as Microsoft Foundry. German is a strong advantage, as many of our clients are in the DACH region. Practical details - Hybrid working and client-site attendance will depend on the engagement. - You will need to complete the background and security checks required by the client. We'd love to hear from you! If your profile looks like a good match we'll be in touch to organise a first stage interview. Please note, due to the high volume of candidates, we cannot offer personalized feedback to each candidate. Find more English Speaking Jobs in United Kingdom on Arbeitnow

About Nebius: Nebius is leading a new era in cloud infrastructure for the global AI economy. We are building a full-stack AI cloud platform that supports developers and enterprises from data and model training through to production deployment, without the cost and complexity of building large in-house AI/ML infrastructure. Built by engineers, for engineers. From large-scale GPU orchestration to inference optimization, we own the hard problems across compute, storage, networking and applied AI. Listed on Nasdaq (NBIS) and headquartered in Amsterdam, we have a global footprint with R&D hubs across Europe, the UK, North America and Israel. Our team of 1,500+ includes hundreds of engineers with deep expertise across hardware, software and AI R&D. The role As a Senior Machine Learning Engineer on the Applied AI team at Nebius Token Factory, you will own model and endpoint optimization from model artifacts through production deployment. Your work will span model internals, quantization and model compression, speculative decoding, KV-cache optimization, inference engines, serving architecture, and benchmarking to improve latency, throughput, memory efficiency, GPU utilization, and cost per token while preserving quality and reliability. You will work with kernel and platform engineers to investigate serving problems, compare configurations, and resolve performance and quality regressions. You will evaluate model- and engine-level optimizations together with distributed inference designs, including request routing, scheduling, prefill/decode coordination, and multi-node GPU execution. Your improvements will be validated through reproducible benchmarks and safe production rollouts under real-world workloads. Your responsibilities Own optimization projects for specific model families, customer endpoints, or serving backends. Compare inference engines and recommend serving configurations suited to individual workloads. Diagnose model-quality and performance regressions during production rollouts. Improve LLM and VLM endpoint latency, throughput, memory efficiency, GPU utilization, quality, and cost per token. Deploy, configure, benchmark, and extend engines such as vLLM, SGLang, TensorRT-LLM, Triton Inference Server, and NVIDIA Dynamo. Develop production model-compression workflows covering quantization, quantization-aware training, distillation, low-bit serving, and accuracy recovery. Implement or integrate speculative decoding, draft-model approaches, KV-cache optimization, prefix caching, chunked prefill, continuous batching, and disaggregated prefill/decode serving. For prefill–decode disaggregation (PDD), evaluate KV-cache transfer, worker placement, and capacity balancing to determine which workloads benefit from the architecture. Design and improve LLM request routers and scheduling policies that balance worker utilization, queueing, request characteristics, and KV-cache locality while meeting latency and reliability targets. Scale dense and mixture-of-experts inference across multiple GPU nodes. Select and tune tensor, pipeline, data, and expert parallelism—including wide expert parallelism (WideEP)—with attention to hardware topology, expert load balance, and communication overhead. Build reproducible benchmark harnesses measuring TTFT, TPOT, tokens per second per GPU, p95/p99 latency, GPU memory, reliability, and cost per token. Validate architecture choices under representative traffic and consistent GPU budgets. Collaborate with GPU kernel and platform engineers to trace bottlenecks across model code, kernels, runtime, scheduler, gateway, and cluster layers. Produce design documents, performance reports, rollout plans, and customer-facing technical explanations. Must-haves Strong engineering skills in Python and PyTorch. Hands-on experience deploying or optimizing LLM, VLM, or high-throughput transformer inference. Practical knowledge of at least one modern inference stack, such as vLLM, SGLang, TensorRT-LLM, Triton Inference Server, NVIDIA Dynamo, Ray Serve, KServe, or an equivalent internal system. A strong understanding of transformer inference bottlenecks involving KV cache, attention, memory bandwidth, batching, parallelism, and long-context serving. Ability to quantify trade-offs among latency, throughput, quality, utilization, and cost. Experience designing or optimizing distributed inference system architecture, with hands-on work in one or more areas such as request routing, distributed scheduling, PDD, multi-node inference, or expert parallelism. Strong communication and collaboration skills across research, kernel, infrastructure, product, and customer teams. Nice-to-haves Experience with quantization-aware training, post-training quantization, FP8, INT8, INT4, NVFP4, MXFP4, AWQ, GPTQ, or SmoothQuant. Experience with distillation, speculative decoding, EAGLE, Medusa, multi-token prediction, or related inference acceleration methods. Experience supporting agentic workloads involving tool calling, structured outputs, streaming APIs, high concurrency, and multi-step orchestration. Familiarity with CUDA or Triton; the role does not require kernel engineering to be the candidate's primary specialization. Contributions to open-source projects such as vLLM, SGLang, TensorRT-LLM, FlashInfer, LMCache, PyTorch, Triton, Ray, or KServe. Implementation experience with cache-aware request routing, PDD, or WideEP, including diagnosing communication bottlenecks and load imbalance. Familiarity with GPU communication libraries and interconnects, such as NCCL, NVLink, InfiniBand, or RoCE. Benefits & Perks: Competitive compensation Career growth and learning opportunities Flexibility and ownership Collaborative and innovative culture Opportunity to work on impactful AI projects International environment and talented teams What's it like to work at Nebius: Fast moving - Bold thinking - Constant growth - Meaningful impact - Trust and real ownership - Opportunity to shape the future of AI Equal Opportunity Statement: Nebius is an equal opportunity employer. We are committed to fostering an inclusive and diverse workplace and to providing equal employment opportunities in all aspects of employment. We do not discriminate on the basis of race, color, religion, sex (including pregnancy), national origin, ancestry, age, disability, genetic information, marital status, veteran status, sexual orientation, gender identity or expression, or any other characteristic protected by applicable law. Applicants must be authorized to work in the country in which they apply and will be required to provide proof of employment eligibility as a condition of hire. If you need accommodations during the application process, please let us know. Find more English Speaking Jobs in Germany on Arbeitnow

About Nebius: Nebius is leading a new era in cloud infrastructure for the global AI economy. We are building a full-stack AI cloud platform that supports developers and enterprises from data and model training through to production deployment, without the cost and complexity of building large in-house AI/ML infrastructure. Built by engineers, for engineers. From large-scale GPU orchestration to inference optimization, we own the hard problems across compute, storage, networking and applied AI. Listed on Nasdaq (NBIS) and headquartered in Amsterdam, we have a global footprint with R&D hubs across Europe, the UK, North America and Israel. Our team of 1,500+ includes hundreds of engineers with deep expertise across hardware, software and AI R&D. Role Summary Nebius Token Factory is building production-grade inference for open-source, customer, and internally optimized models. This role owns the path from model artifact to fast, reliable, cost-efficient endpoint. The work sits at the intersection of model internals, inference engines, serving architecture, benchmarking, and production deployment. What You Will Do Own a platform-level optimization pillar such as quantization, speculative decoding, KV -cache optimization, engine selection, or long-context serving. Create reusable evaluation, benchmarking, and rollout frameworks across multiple model families. Provide technical direction to MLEs, Applied Scientists, GPU kernel engineers, and infrastructure engineers on shared performance goals. Optimize LLM and VLM endpoints for latency, throughput, memory efficiency, GPU utilization, quality, and cost per token. Deploy, configure, benchmark, and extend inference engines such as vLLM, SGLang, TensorRT-LLM, Triton Inference Server, NVIDIA Dynamo, or similar systems. Build and productionize model-compression workflows, including quantization, quantization-aware training, distillation, low-bit serving, and accuracy recovery. Implement or integrate speculative decoding, draft-model approaches, KV -cache optimization, prefix caching, chunked prefill, continuous batching, and disaggregated prefill/decode serving. Build reproducible benchmark harnesses for TTFT , TPOT , tokens per second per GPU, p95/p99 latency, GPU memory, reliability, and cost per token. Partner with GPU kernel engineers and platform engineers to diagnose bottlenecks across model code, kernels, runtime, scheduler, gateway, and cluster layers. Write clear design docs, performance reports, rollout plans, and customer-facing technical explanations. Required Qualifications Strong Python and PyTorch engineering skills. Hands-on experience deploying or optimizing LLM, VLM , or high-throughput transformer inference systems. Practical knowledge of at least one modern inference stack such as vLLM, SGLang, TensorRT-LLM, Triton Inference Server, NVIDIA Dynamo, Ray Serve, KServe, or equivalent internal systems. Strong understanding of transformer inference bottlenecks, including KV cache, attention, memory bandwidth, batching, parallelism, and long-context serving. Ability to reason quantitatively about latency, throughput, quality, utilization, and cost tradeoffs. Strong communication skills and ability to collaborate with research, kernel, infrastructure, product, and customer teams. Preferred Qualifications Experience with quantization-aware training, post-training quantization, FP8 , INT8 , INT4 , NVFP4 , MXFP4 , AWQ , GPTQ , SmoothQuant, or related techniques. Experience with distillation, speculative decoding, EAGLE, Medusa, multi-token prediction, or other inference acceleration methods. Experience with agentic workloads, including tool calling, structured outputs, streaming APIs, high concurrency, and multi-step orchestration. CUDA or Triton familiarity, even if the role is not primarily a kernel-engineering role. Open-source contributions to vLLM, SGLang, TensorRT-LLM, FlashInfer, LMCache, PyTorch, Triton, Ray, KServe, or related projects. Benefits & Perks: Competitive compensation Career growth and learning opportunities Flexibility and ownership Collaborative and innovative culture Opportunity to work on impactful AI projects International environment and talented teams What's it like to work at Nebius: Fast moving - Bold thinking - Constant growth - Meaningful impact - Trust and real ownership - Opportunity to shape the future of AI Equal Opportunity Statement: Nebius is an equal opportunity employer. We are committed to fostering an inclusive and diverse workplace and to providing equal employment opportunities in all aspects of employment. We do not discriminate on the basis of race, color, religion, sex (including pregnancy), national origin, ancestry, age, disability, genetic information, marital status, veteran status, sexual orientation, gender identity or expression, or any other characteristic protected by applicable law. Applicants must be authorized to work in the country in which they apply and will be required to provide proof of employment eligibility as a condition of hire. If you need accommodations during the application process, please let us know. Find more English Speaking Jobs in Germany on Arbeitnow

About Nebius: Nebius is leading a new era in cloud infrastructure for the global AI economy. We are building a full-stack AI cloud platform that supports developers and enterprises from data and model training through to production deployment, without the cost and complexity of building large in-house AI/ML infrastructure. Built by engineers, for engineers. From large-scale GPU orchestration to inference optimization, we own the hard problems across compute, storage, networking and applied AI. Listed on Nasdaq (NBIS) and headquartered in Amsterdam, we have a global footprint with R&D hubs across Europe, the UK, North America and Israel. Our team of 1,500+ includes hundreds of engineers with deep expertise across hardware, software and AI R&D. The role As a Senior Machine Learning Engineer on the Applied AI team at Nebius Token Factory, you will own model and endpoint optimization from model artifacts through production deployment. Your work will span model internals, quantization and model compression, speculative decoding, KV-cache optimization, inference engines, serving architecture, and benchmarking to improve latency, throughput, memory efficiency, GPU utilization, and cost per token while preserving quality and reliability. You will work with kernel and platform engineers to investigate serving problems, compare configurations, and resolve performance and quality regressions. You will evaluate model- and engine-level optimizations together with distributed inference designs, including request routing, scheduling, prefill/decode coordination, and multi-node GPU execution. Your improvements will be validated through reproducible benchmarks and safe production rollouts under real-world workloads. Your responsibilities Own optimization projects for specific model families, customer endpoints, or serving backends. Compare inference engines and recommend serving configurations suited to individual workloads. Diagnose model-quality and performance regressions during production rollouts. Improve LLM and VLM endpoint latency, throughput, memory efficiency, GPU utilization, quality, and cost per token. Deploy, configure, benchmark, and extend engines such as vLLM, SGLang, TensorRT-LLM, Triton Inference Server, and NVIDIA Dynamo. Develop production model-compression workflows covering quantization, quantization-aware training, distillation, low-bit serving, and accuracy recovery. Implement or integrate speculative decoding, draft-model approaches, KV-cache optimization, prefix caching, chunked prefill, continuous batching, and disaggregated prefill/decode serving. For prefill–decode disaggregation (PDD), evaluate KV-cache transfer, worker placement, and capacity balancing to determine which workloads benefit from the architecture. Design and improve LLM request routers and scheduling policies that balance worker utilization, queueing, request characteristics, and KV-cache locality while meeting latency and reliability targets. Scale dense and mixture-of-experts inference across multiple GPU nodes. Select and tune tensor, pipeline, data, and expert parallelism—including wide expert parallelism (WideEP)—with attention to hardware topology, expert load balance, and communication overhead. Build reproducible benchmark harnesses measuring TTFT, TPOT, tokens per second per GPU, p95/p99 latency, GPU memory, reliability, and cost per token. Validate architecture choices under representative traffic and consistent GPU budgets. Collaborate with GPU kernel and platform engineers to trace bottlenecks across model code, kernels, runtime, scheduler, gateway, and cluster layers. Produce design documents, performance reports, rollout plans, and customer-facing technical explanations. Must-haves Strong engineering skills in Python and PyTorch. Hands-on experience deploying or optimizing LLM, VLM, or high-throughput transformer inference. Practical knowledge of at least one modern inference stack, such as vLLM, SGLang, TensorRT-LLM, Triton Inference Server, NVIDIA Dynamo, Ray Serve, KServe, or an equivalent internal system. A strong understanding of transformer inference bottlenecks involving KV cache, attention, memory bandwidth, batching, parallelism, and long-context serving. Ability to quantify trade-offs among latency, throughput, quality, utilization, and cost. Experience designing or optimizing distributed inference system architecture, with hands-on work in one or more areas such as request routing, distributed scheduling, PDD, multi-node inference, or expert parallelism. Strong communication and collaboration skills across research, kernel, infrastructure, product, and customer teams. Nice-to-haves Experience with quantization-aware training, post-training quantization, FP8, INT8, INT4, NVFP4, MXFP4, AWQ, GPTQ, or SmoothQuant. Experience with distillation, speculative decoding, EAGLE, Medusa, multi-token prediction, or related inference acceleration methods. Experience supporting agentic workloads involving tool calling, structured outputs, streaming APIs, high concurrency, and multi-step orchestration. Familiarity with CUDA or Triton; the role does not require kernel engineering to be the candidate's primary specialization. Contributions to open-source projects such as vLLM, SGLang, TensorRT-LLM, FlashInfer, LMCache, PyTorch, Triton, Ray, or KServe. Implementation experience with cache-aware request routing, PDD, or WideEP, including diagnosing communication bottlenecks and load imbalance. Familiarity with GPU communication libraries and interconnects, such as NCCL, NVLink, InfiniBand, or RoCE. Benefits & Perks: Competitive compensation Career growth and learning opportunities Flexibility and ownership Collaborative and innovative culture Opportunity to work on impactful AI projects International environment and talented teams What's it like to work at Nebius: Fast moving - Bold thinking - Constant growth - Meaningful impact - Trust and real ownership - Opportunity to shape the future of AI Equal Opportunity Statement: Nebius is an equal opportunity employer. We are committed to fostering an inclusive and diverse workplace and to providing equal employment opportunities in all aspects of employment. We do not discriminate on the basis of race, color, religion, sex (including pregnancy), national origin, ancestry, age, disability, genetic information, marital status, veteran status, sexual orientation, gender identity or expression, or any other characteristic protected by applicable law. Applicants must be authorized to work in the country in which they apply and will be required to provide proof of employment eligibility as a condition of hire. If you need accommodations during the application process, please let us know. Find more English Speaking Jobs in Netherlands on Arbeitnow

About Nebius: Nebius is leading a new era in cloud infrastructure for the global AI economy. We are building a full-stack AI cloud platform that supports developers and enterprises from data and model training through to production deployment, without the cost and complexity of building large in-house AI/ML infrastructure. Built by engineers, for engineers. From large-scale GPU orchestration to inference optimization, we own the hard problems across compute, storage, networking and applied AI. Listed on Nasdaq (NBIS) and headquartered in Amsterdam, we have a global footprint with R&D hubs across Europe, the UK, North America and Israel. Our team of 1,500+ includes hundreds of engineers with deep expertise across hardware, software and AI R&D. Role Summary Nebius Token Factory is building production-grade inference for open-source, customer, and internally optimized models. This role owns the path from model artifact to fast, reliable, cost-efficient endpoint. The work sits at the intersection of model internals, inference engines, serving architecture, benchmarking, and production deployment. What You Will Do Own a platform-level optimization pillar such as quantization, speculative decoding, KV -cache optimization, engine selection, or long-context serving. Create reusable evaluation, benchmarking, and rollout frameworks across multiple model families. Provide technical direction to MLEs, Applied Scientists, GPU kernel engineers, and infrastructure engineers on shared performance goals. Optimize LLM and VLM endpoints for latency, throughput, memory efficiency, GPU utilization, quality, and cost per token. Deploy, configure, benchmark, and extend inference engines such as vLLM, SGLang, TensorRT-LLM, Triton Inference Server, NVIDIA Dynamo, or similar systems. Build and productionize model-compression workflows, including quantization, quantization-aware training, distillation, low-bit serving, and accuracy recovery. Implement or integrate speculative decoding, draft-model approaches, KV -cache optimization, prefix caching, chunked prefill, continuous batching, and disaggregated prefill/decode serving. Build reproducible benchmark harnesses for TTFT , TPOT , tokens per second per GPU, p95/p99 latency, GPU memory, reliability, and cost per token. Partner with GPU kernel engineers and platform engineers to diagnose bottlenecks across model code, kernels, runtime, scheduler, gateway, and cluster layers. Write clear design docs, performance reports, rollout plans, and customer-facing technical explanations. Required Qualifications Strong Python and PyTorch engineering skills. Hands-on experience deploying or optimizing LLM, VLM , or high-throughput transformer inference systems. Practical knowledge of at least one modern inference stack such as vLLM, SGLang, TensorRT-LLM, Triton Inference Server, NVIDIA Dynamo, Ray Serve, KServe, or equivalent internal systems. Strong understanding of transformer inference bottlenecks, including KV cache, attention, memory bandwidth, batching, parallelism, and long-context serving. Ability to reason quantitatively about latency, throughput, quality, utilization, and cost tradeoffs. Strong communication skills and ability to collaborate with research, kernel, infrastructure, product, and customer teams. Preferred Qualifications Experience with quantization-aware training, post-training quantization, FP8 , INT8 , INT4 , NVFP4 , MXFP4 , AWQ , GPTQ , SmoothQuant, or related techniques. Experience with distillation, speculative decoding, EAGLE, Medusa, multi-token prediction, or other inference acceleration methods. Experience with agentic workloads, including tool calling, structured outputs, streaming APIs, high concurrency, and multi-step orchestration. CUDA or Triton familiarity, even if the role is not primarily a kernel-engineering role. Open-source contributions to vLLM, SGLang, TensorRT-LLM, FlashInfer, LMCache, PyTorch, Triton, Ray, KServe, or related projects. Benefits & Perks: Competitive compensation Career growth and learning opportunities Flexibility and ownership Collaborative and innovative culture Opportunity to work on impactful AI projects International environment and talented teams What's it like to work at Nebius: Fast moving - Bold thinking - Constant growth - Meaningful impact - Trust and real ownership - Opportunity to shape the future of AI Equal Opportunity Statement: Nebius is an equal opportunity employer. We are committed to fostering an inclusive and diverse workplace and to providing equal employment opportunities in all aspects of employment. We do not discriminate on the basis of race, color, religion, sex (including pregnancy), national origin, ancestry, age, disability, genetic information, marital status, veteran status, sexual orientation, gender identity or expression, or any other characteristic protected by applicable law. Applicants must be authorized to work in the country in which they apply and will be required to provide proof of employment eligibility as a condition of hire. If you need accommodations during the application process, please let us know. Find Jobs in Netherlands on Arbeitnow

About Nebius: Nebius is leading a new era in cloud infrastructure for the global AI economy. We are building a full-stack AI cloud platform that supports developers and enterprises from data and model training through to production deployment, without the cost and complexity of building large in-house AI/ML infrastructure. Built by engineers, for engineers. From large-scale GPU orchestration to inference optimization, we own the hard problems across compute, storage, networking and applied AI. Listed on Nasdaq (NBIS) and headquartered in Amsterdam, we have a global footprint with R&D hubs across Europe, the UK, North America and Israel. Our team of 1,500+ includes hundreds of engineers with deep expertise across hardware, software and AI R&D. Role Summary Nebius Token Factory is building production-grade inference for open-source, customer, and internally optimized models. This role owns the path from model artifact to fast, reliable, cost-efficient endpoint. The work sits at the intersection of model internals, inference engines, serving architecture, benchmarking, and production deployment. What You Will Do Own a platform-level optimization pillar such as quantization, speculative decoding, KV -cache optimization, engine selection, or long-context serving. Create reusable evaluation, benchmarking, and rollout frameworks across multiple model families. Provide technical direction to MLEs, Applied Scientists, GPU kernel engineers, and infrastructure engineers on shared performance goals. Optimize LLM and VLM endpoints for latency, throughput, memory efficiency, GPU utilization, quality, and cost per token. Deploy, configure, benchmark, and extend inference engines such as vLLM, SGLang, TensorRT-LLM, Triton Inference Server, NVIDIA Dynamo, or similar systems. Build and productionize model-compression workflows, including quantization, quantization-aware training, distillation, low-bit serving, and accuracy recovery. Implement or integrate speculative decoding, draft-model approaches, KV -cache optimization, prefix caching, chunked prefill, continuous batching, and disaggregated prefill/decode serving. Build reproducible benchmark harnesses for TTFT , TPOT , tokens per second per GPU, p95/p99 latency, GPU memory, reliability, and cost per token. Partner with GPU kernel engineers and platform engineers to diagnose bottlenecks across model code, kernels, runtime, scheduler, gateway, and cluster layers. Write clear design docs, performance reports, rollout plans, and customer-facing technical explanations. Required Qualifications Strong Python and PyTorch engineering skills. Hands-on experience deploying or optimizing LLM, VLM , or high-throughput transformer inference systems. Practical knowledge of at least one modern inference stack such as vLLM, SGLang, TensorRT-LLM, Triton Inference Server, NVIDIA Dynamo, Ray Serve, KServe, or equivalent internal systems. Strong understanding of transformer inference bottlenecks, including KV cache, attention, memory bandwidth, batching, parallelism, and long-context serving. Ability to reason quantitatively about latency, throughput, quality, utilization, and cost tradeoffs. Strong communication skills and ability to collaborate with research, kernel, infrastructure, product, and customer teams. Preferred Qualifications Experience with quantization-aware training, post-training quantization, FP8 , INT8 , INT4 , NVFP4 , MXFP4 , AWQ , GPTQ , SmoothQuant, or related techniques. Experience with distillation, speculative decoding, EAGLE, Medusa, multi-token prediction, or other inference acceleration methods. Experience with agentic workloads, including tool calling, structured outputs, streaming APIs, high concurrency, and multi-step orchestration. CUDA or Triton familiarity, even if the role is not primarily a kernel-engineering role. Open-source contributions to vLLM, SGLang, TensorRT-LLM, FlashInfer, LMCache, PyTorch, Triton, Ray, KServe, or related projects. Benefits & Perks: Competitive compensation Career growth and learning opportunities Flexibility and ownership Collaborative and innovative culture Opportunity to work on impactful AI projects International environment and talented teams What's it like to work at Nebius: Fast moving - Bold thinking - Constant growth - Meaningful impact - Trust and real ownership - Opportunity to shape the future of AI Equal Opportunity Statement: Nebius is an equal opportunity employer. We are committed to fostering an inclusive and diverse workplace and to providing equal employment opportunities in all aspects of employment. We do not discriminate on the basis of race, color, religion, sex (including pregnancy), national origin, ancestry, age, disability, genetic information, marital status, veteran status, sexual orientation, gender identity or expression, or any other characteristic protected by applicable law. Applicants must be authorized to work in the country in which they apply and will be required to provide proof of employment eligibility as a condition of hire. If you need accommodations during the application process, please let us know. Find Jobs in Switzerland on Arbeitnow

About us EggAI is a fast-growing AI consulting and development startup. We build agentic AI solutions for enterprise clients. We are growing in several European markets. The role You will own the design of a shared AI platform for our clients. The platform has two parts. An AI Control Plane governs what agents may do, through identity and delegated permissions, policy enforcement, approvals, budgets and durable audit records. A managed AI Runtime is where the agents run. Both integrate with the client's identity, security, delivery and audit services. You will set the architecture, define how services and controls fit together, and make sure what the team builds is sound, assured and ready for production. The role reaches across engineering, security, audit, risk and the client's architecture function. You will take designs through their governance, record the decisions, and give the engineers a design they can build and the client a design they can approve. You will be EggAI's accountable architecture lead on the engagement, leading design for a delivery team of platform engineers. You make the technical decisions within the team's remit, recommend on the rest, and secure the client's approval where their architecture or risk governance requires it. What you will do - Own the architecture of the AI Control Plane and AI Runtime and keep it aligned with the client's architecture standards, cloud substrate, API gateway, identity provider and audit services. - Define service interfaces and contracts. Decide where each security and policy control is enforced: in the gateway, the control plane, the runtime or the client's own services. - Keep the design record. Write architecture decision records (ADRs) and record every deviation from the client's standards or our reference designs for a client decision. - Take designs through the client's architecture and risk forums. Secure approvals in time for dependent work, so that design precedes implementation. - Lead design and quality review across all platform services. Review engineers' changes, cross-service tests and fault exercises with the CTO. - Define how the architecture will be proven. Agree acceptance criteria with engineers and client control owners for each interface, control and infrastructure contract, including behaviour when a dependency fails, and review the test evidence showing they are met. - Hold the build to the design's principles: policy enforced outside the agent, decisions recorded before actions take effect, fail closed with stop always available, and audit evidence that is administered independently of the people who build and run the platform, reconciled against what actually happened, and readable with the client's own tools. - Maintain the inventory of platform controls. Record for each one whether it is enforced or only monitored, whether it is implemented, tested or still open, and what evidence it produces, so acceptance reviews work from a current record and coverage gaps are visible. - Assess proposed technology, scope and dependency changes for their effect on the design, controls, effort and running costs, and give the programme lead a clear basis for plan and cost decisions. - Evaluate new runtimes and managed services against the platform's control requirements. Compare shared controls with provider-native ones on integration effort, operating cost and assurance gaps, separate what is verified from what is assumed, and recommend whether to adopt, change the design or hold scope. - Lead the technical assessment of production readiness. Review recovery, containment and fault-exercise evidence, and give the programme lead a clear view of what is ready and what is not. - Stay with the platform through go-live and the support period after it. Resolve the architectural issues that production exposes, and decide when a fix is a design change rather than a defect. - Prepare the client's receiving engineers to maintain the design. Lead the architecture parts of training and handover, and leave documentation, decision records and guidance they can extend without you. Essential experience - You have been the accountable solution or platform architect for systems that went into production, and have led engineers through building them. - You have delivered cloud platforms into production in a regulated environment and taken designs through formal architecture and risk governance. - You know a major cloud platform in depth: networking, identity, secrets, Kubernetes and managed services. Azure is strongly preferred. - You have hands-on experience with API gateways, identity and delegation using OAuth and OIDC, and policy as code. - You work with infrastructure as code, GitOps and software supply-chain controls, and understand how they underpin reproducible environments, conformance tests and release assurance. - You have designed systems that must produce audit evidence and fail safely: durable records, immutable storage, separation of duties and deny-by-default access. You have sound judgement about retries, partial failure, idempotency, identity propagation and recovery. - You write clear architecture decision records and can explain and defend a design to senior architects, risk owners and engineers. Nice to have - Delivery in financial services or banking. - LLM or agent platforms in production, including model gateways, agent orchestration and protocols such as MCP and A2A. - Audit, evidence retention or AI risk controls, including EU AI Act, PRA or FCA expectations. - Products in this space such as Kong, Open Policy Agent or Microsoft Foundry. Practical details - Hybrid working and client-site attendance will depend on the engagement. - You will need to complete the background and security checks required by the client. Find more English Speaking Jobs in France on Arbeitnow