Abhilash Anuku

I'm a master's student in Applied Computer Science at SRH University Stuttgart, finishing in September 2027. I picked Germany because the applied AI work I want to do — systems that ship into real products, not slide decks — happens here. The plan is simple: build, finish the degree, and stay on a team that's shipping.

A recent project pulled me deep into this side of the field: Mörbius — a multi-agent medical AI that explores diagnostic reasoning across 532 PubMed case reports. Six agents cascade across NVIDIA NIM, Anthropic, HuggingFace and a local Ollama fallback, with a knowledge-graph layer that catches what plain RAG misses. 588 of 588 tests are green. The whole thing runs on free tiers — a zero-budget global deploy, because a system that only works on someone else's credit card is a demo, not a system. Building it taught me how AI gets stitched into the surrounding pieces that make a system actually run — orchestration, failover, evaluation, the unglamorous glue. Alongside Mörbius I'm building LISA — my own self-hosted, MCP-native personal AI with four capability servers and a five-provider failover chain. Both are the same bet from two sides: applied AI that ships, runs, and survives a power cut.

Outside that work I'm reading through quantum computing for computer scientists the long way — one page at a time. Off the keyboard I cook, take photographs, travel when I can, and show up to design-thinking workshops in Stuttgart to meet people who build things you can hold.