Senior LLM Developer for the Neurolegal Service

Senior LLM Developer for the Neurolegal Service

Neurolegal has already become one of the best solutions on the market and is now actively growing and scaling. We are expanding the team and looking for a strong ML developer with experience in RAG, LLM, and agents.

What tasks await you

Improvement of LLM and RAG pipelines Improve LLM and RAG pipelines: optimization of each RAG component, configuration of embedder/reranker models, prompt engineering, generation verifiability (grounding).

Development of Legal LLM Agents Design, develop, implement LLM agents and evaluate their performance for complex scenarios: analysis of user task, planning, selection of appropriate tools at each iteration of plan execution.

Fine-tuning and Scaling Fine-tune LLM/ML models, including Reinforcement Learning from Human Feedback (RLHF) in collaboration with Yandex's core technology teams. Scale production solutions: latency, scalability, model monitoring.

More about ML at Yandex — in the channel Yandex for ML

We expect you to have

• Over two years of experience in ML/NLP/LLM projects, including production systems
• Write code (Python)
• Have experience with LLM, RAG
• Are interested in agents
• Are ready to deploy ML models in runtime, support them, and automate related processes

Will be a plus

• Have performed fine-tuning of complex RAG pipelines
• Have developed LLM agents (LangGraph, AutoGen, Pydantic AI, etc.)
• Have participated in the development of dialog services
• Understand the legal field