Description
We are the R&D team of the GigaLegal project at SBER, creating solutions in the legal domain based on LLM for automating legal processes.
Our goal is to transform the work of lawyers, businesses, and government structures through:
Automation of contract and legal document analysis
Intelligent generation of legal texts
Development of specialized legal chatbots
Creation of autonomous legal AI agents
We are looking for an experienced TL|Senior specialist who will take responsibility for the development and implementation of LLM solutions capable of scaling our clients' business processes.
Responsibilities
LLM-oriented solutions:
- design and implementation of LLM adaptation strategies (prompting, fine-tuning, LoRA, RLHF) for the specifics of legal tasks
- design and development of pipelines for processing legal data (RAG, agent systems, semantic search)
- design of skills and training of LLM and NLP/Classic ML models to implement business tasks
- optimization of LLM performance in production (latency, cost, accuracy).
Production engineering:
- deployment of DS models to production using MLOps practices (CI/CD, monitoring, A/B testing)
- integration of solutions with external APIs, working with vector databases, search engines (ElasticSearch)
- design of fault-tolerant systems for processing confidential legal data
- work with SQL/NoSQL databases
Leadership and expertise:
- participation in setting technical requirements and interaction with business customers
- elaboration of requirements and solution options with legal experts, system analysts, and the customer's side
- elaboration of test and training data labeling for training legal skills of GigaChat and other LLMs with the training department
- mentoring junior colleagues, code review, development of best practices for the team
- risk analysis and finding compromises between model quality, speed, and cost
- possible leadership of the project's RND team and TeamLead position.
Requirements
- higher education
- 5+ years of experience in DS/NLP, including at least 1 year of work with LLMs, production experience.
- willingness to both write code, pipelines, train models, and write documentation, design systems, and prepare specifications for models, data, pipelines
- deep expertise in LLM adaptation: SFT, RLHF, LoRA, prompt engineering
- experience building RAG systems, agent pipelines, and services based on LLMs
- knowledge of modern frameworks (PyTorch, Hugging Face, LangChain, LlamaIndex)
- confident work with infrastructure: Docker, Kubernetes, cloud platforms
- understanding of MLOps: CI/CD, data drift monitoring, logging
- experience transforming business tasks into technical requirements
- ability to evaluate the ROI of DS solutions and balance between innovation and practicality
- ability to quickly prototype solutions and find a balance between speed/quality/performance.
Conditions
- comfortable modern office
- hybrid work format
- annual salary review, annual bonus
- corporate gym and recreation areas
- more than 400 educational programs at SberUniversity for professional and career development
- extended DMS (voluntary health insurance), preferential insurance for family, and corporate pension program
- employee mortgage with a benefit of up to 7%
- free SberPrime+ subscription, discounts on products from partner companies
- referral bonus for recommending friends to the Sber team.
