Masterand - Hybride Inference API (all gender)
Hybrid
Core
Develop a hybrid inference API for coding agents that routes tasks between local and cloud-based language models.
Role type
Master's thesis research project (IC)
Builds
A prototype inference API for GitHub Copilot CLI
Domain
AI/LLM development and API engineering
Deliverable
production ML models
Required skills
Python, REST APIs, FastAPI, LLM inference, Model routing, RAG, Local LLMs
Preferred skills
Coding agent development, Streaming protocols, Embeddings
Responsibilities
Analyze requirements for OpenAI-compatible APIs, Design architecture for model routing, Implement a prototype, Evaluate routing rules and response quality, Document results and recommendations
