christoph@berlin: ~

$ whoami

Christoph Stach

Senior Software Engineer//Berlin//12+ yrs

$ cat about.txt

Senior software engineer with a strong background in web development who owned features across the full development cycle, enforced code quality and kept software maintainable. Built LLM agents and worked with AI coding tools. Experience across enterprise, mid-sized and startup environments.

Portrait of Christoph Stach
portrait.webp, 1 image rendered
  • 12+years_exp
  • 3companies
  • 2cs_degrees
  • 36technologies

At MBition (Mercedes-Benz), I work as a Senior Software Engineer and own the Python platform on which multiple Mercedes-Benz teams build and run their LLM agents on Kubernetes. Its service is the cloud backend for the in-car voice assistant serving several million vehicles. I re-architected it to fully asynchronous Python, which cut its memory use by 50%, and rebuilt the CI/CD pipelines, which took deployment from a ~2-hour manual process to an automated 15 minutes.

Before that, on the company's speech platform, I built a no-code platform for configuring and deploying speech applications, shipping features end-to-end in Vue and Nuxt with strict TypeScript. I owned code quality across the team and cut build times by ~50% by consolidating multiple products into a pnpm monorepo.

Earlier, alongside my studies at DERICON, I built and owned a graph-based no-code workflow builder in React and Remix. At SLH, I modernized a legacy PHP application into a modular frontend/backend architecture.

I hold a B.Sc. and an M.Sc. in Applied Computer Science from HTW Berlin, specializing in data science and health computing, with theses on style transfer for low-power hardware and nanopore-based DNA sequencing.

  • frontend[10]
    • React
    • Next.js
    • Vue
    • Nuxt
    • Remix
    • TypeScript
    • JavaScript
    • Tailwind
    • TanStack Query
    • React Server Components
  • backend_api[6]
    • Python
    • FastAPI
    • Node.js
    • NestJS
    • REST APIs
    • zod
  • data[4]
    • PostgreSQL
    • MongoDB
    • Prisma
    • Drizzle
  • ai_llm[6]
    • LLMs
    • AI Agents
    • Model Context Protocol (MCP)
    • RAG
    • LangChain
    • Prompt Engineering
  • infra_cicd[4]
    • Docker
    • Kubernetes
    • GitLab CI
    • pnpm Monorepos
  • practices[6]
    • AI-Assisted Development (Claude Code, GitHub Copilot)
    • Unit Testing (Vitest, Jest, pytest)
    • Dependency Injection
    • Code Review
    • Agile
    • Scrum