Suggested rewrite: Led a cross-functional initiative that improved [business outcome] by [measurable result], demonstrating experience relevant to this role...
Senior ML Engineer (AI Research)
Review the role, location requirements, compensation details, and application process before deciding whether this opportunity fits your next career move.
- Salary
- Undisclosed
- Department
- Data Science & Analytics
- Employment
- Full Time
- Experience
- Senior
- Published
- Apply before
- 26 Nov 2026
- Listing views
- 23
- Application actions
- 2
Make your next move.
Prepare your resume, explore your fit, and draft a cover letter for this opportunity.
The role, at a glance.
Nebius is seeking a senior or staff-level machine learning engineer for applied AI research focused on agentic systems, language models, and reinforcement learning. The role involves designing statistically rigorous experiments, training large models across multiple compute nodes, and translating research findings into practical collaboration with adjacent engineering teams. Key research areas include guided search, reinforcement learning for reasoning, web-scale data collection, LLM post-training, and model distillation. Candidates need deep ML and RL theory, strong Python and JAX engineering skills, distributed-training experience, and the ability to communicate research results clearly. This is a high-impact role in a fast-moving AI infrastructure and research environment.
Role DNA
A quick view of the complexity, pace, ownership and collaboration implied by the job description.
Pace & Pressure
5/5Autonomy Level
5/5Communication Load
5/5Salary analysis
Estimated compensation compared with the broader US market for similar roles.
Core skills
Skills and capabilities most closely associated with this opportunity.
Sample interview questions
I would first define a clear task distribution, success metric, compute budget, and baseline supervised model. I would control random seeds and major hyperparameters, run repeated trials, and report confidence intervals alongside aggregate reward and task-success metrics. I would then analyze failure modes, reward hacking, training stability, and generalization to held-out tasks before selecting a configuration.
Major challenges include communication overhead, optimizer-state memory, fault tolerance, data-loading bottlenecks, and reproducibility. I would use appropriate parallelism strategies, such as data, tensor, or pipeline parallelism; memory-efficient sharding such as FSDP or ZeRO; profiling to identify bottlenecks; and robust checkpointing. I would also ensure deterministic experiment configuration and detailed tracking of infrastructure and training metrics.
I would begin with a precise definition of useful agent trajectories and target task types. The collection pipeline would include source-quality filters, deduplication, licensing and privacy review, contamination checks, repository and issue-context extraction, and automated validation of task solvability. I would continuously evaluate data quality using downstream agent performance and targeted human or programmatic audits.
PPO directly optimizes a reward signal and can support complex online interaction settings, but it is operationally more complex and sensitive to reward-model quality and optimization stability. DPO is typically simpler because it learns directly from preference pairs without an explicit online RL loop, making it efficient for offline preference optimization. The preferred method depends on the availability of reliable preference data, the need for environment interaction, and the desired level of control over reward optimization.
I document the research question, assumptions, datasets, model configurations, code versions, evaluation protocol, and statistical results from the start. I provide concise experiment reports with charts, ablation results, limitations, and a recommendation that distinguishes evidence from hypothesis. For handoff, I create reproducible training scripts, pinned environments, test coverage, and clear success criteria for production validation.
About this role.
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
This role is for Nebius AI R&D, a team focused on applied research in AI. Examples of applied research that we have recently published include:
- applying reinforcement learning for agent training in long-context multi-turn scenarios
- dramatically scaling task data collection to power reinforcement learning for SWE agents
- building a decontaminated evaluation for SWE agents that is regularly updated
- investigating how test-time guided search can be used to build more powerful agents
The results often lead to collaboration with adjacent teams where our research findings are applied in practice.
We are currently looking for senior- and staff-level ML engineers to work on research in areas such as:
- Guided search and reinforcement learning for agentic systems
- Reinforcement learning for reasoning models
- Web-scale problem collection for training agents
- Efficient model distillation
Some examples of what your responsibilities might include are:
- Conducting experiments to figure out efficient ways to train a large language model on traces of interactions with various environments
- Exploring methods of guided generation and search in the trajectory space
- Coming up with ways to mine relevant data at web scale and figuring out efficient ways to use this data in model post-training
- Conducting experiments with different reinforcement learning configurations in verifiable domains
- Exploring methods to train AI agents on tasks with non-verifiable reward signals
We expect you to have:
- A profound understanding of theoretical foundations of machine learning and reinforcement learning
- Deep expertise in modern deep learning for language processing and generation
- Substantial experience with training large models on multiple computational nodes
- Strong software engineering skills (we mostly use python)
- Deep experience with modern deep learning frameworks (we use jax)
- Strong communication and leadership abilities
- Experience designing, executing, and analyzing machine learning experiments with proper statistical rigor
- Ability to formulate research questions, design experiments to test hypotheses, and draw meaningful conclusions from results
- Ability to document research findings clearly and contribute to technical publications or report
Nice to have:
- Experience with deep reinforcement learning for LLMs, including techniques such as reward modeling, DPO, PPO etc
- Familiarity with important ideas in LLM space, such as RoPE, ZeRO/FSDP, Flash Attention, quantization
- Bachelor’s degree in Computer Science, Artificial Intelligence, Data Science, or a related field Master’s or PhD preferred
- Track record of building and delivering products (not necessarily ML-related) in a dynamic startup-like environment
- Experience in engineering complex systems, such as large distributed data processing systems or high-load web services
- Open-source projects that showcase your engineering prowess
- Excellent command of the English language, alongside superior writing, articulation, and communication skills
- Proficiency in contemporary software engineering approaches, including CI/CD, version control and unit testing
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.
Annual salary information is not provided for this position. Explore salary ranges for similar roles in our Salary Directory ›
This job listing has been manually reviewed by the Jobicy Trust & Safety Team for compliance with our posting guidelines, including verification of the company's legitimacy, accuracy of job details, clarity of remote work policy, and absence of misleading or fraudulent content.
Apply now.
Follow the employer’s application method and review Jobicy’s safety guidance before sharing personal information.
Continue on the employer website
Protect your personal information and never pay to secure an interview or job offer. .
