Python Developer for AI Services (LLM / RAG)

TEAMLY is an IT platform for collaboration and knowledge management. It seamlessly integrates team collaboration, knowledge base management, and corporate training functions.

With its diverse functionality, it allows you to translate all of a company's accumulated experience into practical skills and knowledge for employees.

Well-known companies trust us with their knowledge: VkusVill, Analytical Center under the Government of the Russian Federation, Capital Life, 1C-Rarus, Interleasing. As well as medium and small businesses.

Currently, the TEAMLY team is actively growing, which is why we would be happy to welcome an AI Developer to our team. A sought-after product guarantees rapid growth and interesting tasks.

• Developing HTTP APIs and business logic for two asynchronous Python services using FastAPI
• Designing and implementing RAG pipelines: retrieval, query rewriting, multi-hop search, context aggregation, structured output, source citation, and handling partial responses
• Developing agent scenarios using LangChain, LangGraph, and deepagents: tools, middleware, call limits, cycle protection, and response contract control
• Improving semantic search quality: chunking, dense/sparse retrieval, Qdrant filters, reranking, RRF, and token budget
• Developing event-driven indexing for articles, files, and structured content via Kafka
• Ensuring correct multi-tenancy and filtering of results by user, group, and community rights
• Integrating and replacing LLM, embedding, and reranking providers, including OpenAI-compatible API, vLLM, and Ollama
• Working with PostgreSQL, Redis, and Qdrant, designing data models and Alembic migrations
• Diagnosing quality, performance, and reliability issues at the boundaries of external services
• Developing automated tests and quality gates, maintaining reproducible local and CI environments
• Enhancing observability: structured logs, technical metrics, OpenTelemetry, and LLM/RAG tracing via Langfuse
• Participating in task decomposition, code reviews, and architectural decision-making

What is important to us

• Python 3.12, async/await;

• FastAPI, Pydantic 2;

• SQLAlchemy 2 async, Alembic, PostgreSQL;

• Redis, Apache Kafka, aiokafka;

• Qdrant, dense/sparse vectors, embeddings, reranking, RRF;

• LangChain, LangGraph, deepagents;

• OpenAI-compatible API, vLLM, Ollama;

• Pytest, pytest-asyncio, mypy, Black, Flake8;

• Docker, Docker Compose, Kubernetes, GitLab CI;

• Langfuse, OpenTelemetry

• Confident proficiency in Python and practical experience in backend service development

• Good understanding of asynchronous programming: event loop, concurrent I/O, timeouts, cancellation, resource management, and graceful shutdown

• Experience developing APIs with FastAPI or a comparable Python framework

• Practical experience with PostgreSQL and SQLAlchemy: transactions, concurrent access, indexes, migrations, and query diagnostics

• Experience integrating external services and designing resilient interaction: retry/backoff, timeout, idempotency, concurrency limiting, and partial failure handling

• Understanding of event-driven architecture and Kafka consumer semantics: consumer groups, offset commit, rebalance, redelivery, and idempotent processing

• Understanding of RAG architecture: chunking, embeddings, retrieval, reranking, context window, LLM limitations, and methods for reducing hallucinations

• Ability to write unit and integration tests for asynchronous code, separating business logic from external infrastructure

• Ability to understand existing codebase, identify root causes of problems, and propose minimal, testable changes

• Office-based, hybrid, or remote work format
• Development – we pay for external courses, offer a corporate discount on English language learning, support participation in conferences, meetups, and webinars. We want to foster the expertise of each employee!
• A large and truly interesting product that changes the world for the better. We use innovative technologies, create cutting-edge solutions, and offer a unique opportunity to be part of a team where your ideas and skills can impact the lives of many people;
• Comfort – we have a great office in the center of Moscow (Kitay-gorod, Kurskaya, Chistye Prudy) with cozy workspaces. We provide everything necessary for work: good hardware, additional monitors, licensed software, and paid access;
• Team – TEAMLY has a truly cohesive team of professionals. We value the work of each employee, support initiatives, and help with problem-solving;
• Vibrant life outside of work – every month we organize corporate events, go out into nature together, visit bars and quizzes, participate in sports events, play airsoft, have movie nights, and watch shows in the recreation room. We strive to make not only work comfortable but also to make leisure enjoyable together!