Description

A new system for automating the Bank's procurement processes is being developed in Java using AI. The system involves the development of an innovative software solution designed to optimize and simplify the procurement process in the banking sector. A key feature of the project is the integration of artificial intelligence methods, such as machine learning and natural language processing, to automate various stages of the procurement process. The system will support automatic document generation, contract management, and integration with existing banking systems.

Responsibilities

• working on matching, search, and entity extraction models (including the use of LLMs)
• fine-tuning generative LLMs for various tasks (including summarization, QA, dialogue generation).

Requirements

• strong knowledge of Python (3.10+)
• knowledge of approaches for evaluating quality and calculating metrics for LLM and NLP solutions;
• basic knowledge of classical NLP methods, understanding of DL architectures;
• commercial development experience in building models for solving NLP/LLM tasks for at least 1 year;
• experience fine-tuning BERT-like models;
• knowledge of the internal workings of LLMs (Transformer);
• experience using LangChain, LangGraph in developing LLM solutions.

Will be a plus:

• knowledge of NLP technologies
• experience in developing or fine-tuning NLP and LLM models.
• knowledge of metrics for matching, search, entity extraction, and summarization tasks.

Conditions

·     comfortable modern office

·     work schedule: office-based

·     annual salary review, quarterly and annual bonus

·     corporate gym and relaxation areas

·     over 400 educational programs from SberUniversity for professional and career development

·     extended voluntary health insurance, preferential family insurance, and corporate pension program

·     flexible mortgage discount equal to 1/3 of the Central Bank's key rate

·     free SberPrime+ subscription, discounts on products from partner companies

·     opportunity to participate in corporate life: "Green Marathon", Sberbankiada, KVN, volunteer programs, etc.

·     referral bonus for recommending friends to the Sber team.