Description
The primary focus of our team is the development of a multi-agent approach to collateral assessment in the process of corporate lending. Our goal is to eliminate human involvement from the assessment process where AI can handle it.
We expect the candidate to implement agents "end-to-end": from identifying the task to acceptance in the production environment with subsequent maintenance of continuous and uninterrupted operation.
Responsibilities
- development and integration of AI agents in Python using modern frameworks (FastAPI, LangGraph, aiohttp, etc.)
- requirement analysis, functional formalization, designing the architecture of agent and service interaction
- elaboration of data sources for agents, configuration of data flows and API integrations
- creation and maintenance of high-quality technical documentation, specifications, ER diagrams, and OpenAPI descriptions
- participation in solution design and code review alongside developers
- collaboration with product and analytics teams to build correct agent application scenarios
- support of the solution lifecycle — from MVP assembly to operation in a production environment
- conducting presentations and technical demos for teams and customers.
Requirements
- solid knowledge of Python and experience creating production services with FastAPI / aiohttp / LangGraph
- experience in designing and launching AI- or rule-based agents in real products
- understanding of principles of asynchronous architecture, microservice integration, REST API, and event-driven approaches
- proficiency with Swagger (OpenAPI), Jira, Confluence, CI/CD tooling, and containerization (Docker)
- knowledge of data design principles, experience building ER models
- developed soft skills: systems thinking, communication skills, ability to find solutions without unnecessary bureaucracy and to bring tasks to completion.
Will be an advantage
- experience in the banking or fintech sector, understanding of credit processes and collateral assessment
- skills in analytical modeling (BPMN, UML) and interaction with business teams.
Conditions
- comfortable Class A office near Kutuzovskaya metro station (hybrid work format)
- opportunity to participate in internal and external IT conferences, a wide selection of training courses in the Corporate University
- annual salary review and an annual bonus
- extended voluntary health insurance (VHI), preferential insurance for family, and a corporate pension program
- free SberPrime+ subscription, discounts on products from partner companies
- referral bonus for recommending friends to the Sber team.
