Research Engineer – AI Agents & LLM Systems F/M
Summary
Research engineer on an AI Agent Research team in Paris, hired via Adoc Talent Management for a global tech company. Day to day is ~70–80% hands-on engineering and 20–30% research: building, evaluating and iterating on LLM-based agent systems using tool use, planning, memory and retrieval, primarily in Python, TypeScript and PyTorch.
Company
Adoc Talent Management is recruiting a Research Engineer – AI Agents & LLM Systems (F/M) for its client, a global technology company expanding its AI research activities in Europe. You will join the AI Agent Research team in Paris, working at the forefront of a rapidly evolving field to explore, build, and evaluate next-generation AI agent systems.
In a rapidly evolving field, the role combines research, experimentation and hands-on software engineering, with a strong focus on building prototypes, testing ideas and developing solutions that can ultimately create value for customers.
Position
As a Research Engineer – AI Agents & LLM Systems (F/M), you will contribute to the development and experimentation of AI-agent systems capable of reasoning, planning, interacting with tools and completing complex tasks.
Your work will involve approaches such as tool use, planning, memory, retrieval, code execution, computer use and multi-step workflows.
The role is primarily hands-on, with approximately 70–80% focused on engineering and building and 20–30% on research and exploration. You will take ownership of technical problems, work with limited supervision, rapidly develop prototypes, evaluate their performance and iterate based on the results.
The problems to be addressed are generally defined according to business needs and the evolving AI landscape, while you will have significant autonomy regarding the technical approach used to solve them.
Key Responsibilities
- AI Agent Development: Design and implement AI-agent systems based on large language models and multimodal foundation models.
- Rapid Prototyping: Build and test prototypes from initial ideas, exploring different technical approaches and iterating rapidly.
- Research & Experimentation: Follow recent research, reproduce relevant approaches, critically assess them and adapt ideas to develop new solutions.
- System Evaluation: Evaluate the systems developed, measure their effectiveness, analyse failures and use the results to improve them.
- Software Engineering: Develop clean and reusable research infrastructure, APIs and internal demonstrations.
- End-to-End Ownership: Take ownership of technical subjects from understanding the problem through implementation, experimentation, evaluation and development towards a working solution.
- Customer Feedback: Where relevant, interact with customers, gather feedback and use it to improve the solutions developed.
- Collaboration: Work closely with researchers and engineers within the team while independently owning substantial technical work.
Profile
You hold a Master's or PhD in Computer Science, Artificial Intelligence, Machine Learning or a related field. You have strong programming and software-engineering skills, particularly in Python, and are comfortable with system design and clean coding. The team currently uses Python and TypeScript.
You have experience building and delivering non-trivial technical, AI or machine-learning systems, with a practical understanding of large language models and modern generative AI systems.
You are comfortable with PyTorch, designing experiments, analysing results and iterating on technical solutions.
Most importantly, you have demonstrated the ability to take ownership of a technical problem and deliver a concrete result, ideally having built something from 0 to 1.
You are curious, autonomous and comfortable with uncertainty, with the ability to move quickly from an idea to implementation and experimentation.
The team is particularly interested in candidates who demonstrate ownership, initiative, curiosity, ambition, independent thinking, collaboration and the ability to execute.
Preferred Qualifications
· Experience building AI agents, LLM systems, coding agents, tool-using systems or other multi-step AI systems
· Experience with LLM evaluation, trajectory analysis, reinforcement learning or synthetic-data generation
· Experience serving or optimising open-source language models
· Experience taking a research or technical project from prototype towards a solution used by customers or users
· Contributions to open-source AI projects
· Publications at recognised AI or systems venues
Ready to join the mission?
Apply now by sending your CV, technical background and relevant projects or publications, if applicable, to Adoc Talent Management.