Paid internship in Sber's B2B AI transformation team. Over 3 months, gain in-depth knowledge of designing LLM-based agent systems with the opportunity to receive an offer for a junior position.
Tasks:
- Study LLM agent design patterns
- Master the LangChain, LangGraph, and internal AEF SDK stack
- Design LLM-based AI agent architecture
- Develop agent systems: ReAct, Plan-and-Execute, Multi-Agent
- Integrate agents with internal APIs and banking systems
- Develop and improve RAG pipelines
- Participate in the implementation of agent solutions in an industrial environment
- Evaluate agent quality: metrics, testing, reliability, behavioral stability
Requirements:
- You are in your final year of a Master's program, have already completed a Master's, or are pursuing a PhD
- You have a strong mathematical, technical, or engineering background
- You are proficient in Python
- You can quickly learn new libraries, frameworks, and approaches
- You are ready to dive deep into solution architecture
Conditions:
- Paid internship
- Work on real business tasks
- In-depth dive into LLM agent and agent system design
- Access to internal stack and tools
- Unlimited GigaCode developer agent tariff
- Opportunity to receive an offer for a Junior Data Scientist / Junior AI Engineer position upon completion of the internship
Skills:
- Python
- LangChain
- LangGraph
- AEF SDK
- RAG
- FAISS
- Chroma
- Qdrant
- Llama
- Mistral
- Qwen
- MLP
- CNN
- RNN
- LSTM
- Transformers
- ML
- MLOps
- DevOps
- API
- CI/CD
- Containerization
