Website:
titlewize.com
Job details:
The role
We're looking for an AI engineer to own and improve our document intelligence pipeline — the core of what we do. This isn't a "build a chatbot" role. You'll be working with messy, multilingual, real-world documents and making sure our AI agents are producing reliable, structured output for our existing and future clients.
What you'll work on
Evaluate and improve our LLM-agents Read prompts and output schemas for proper optimization Improve pipeline reliability across inputs Build and run evals to systematically measure agent output quality Work with FastAPI, Langgraph, Celery, Pydantic
Who you are
You've built or debugged an LLM pipeline in a real project — not just a tutorial You can read a prompt cold and immediately spot where it'll break You care about structured outputs and hate when an LLM makes something up Messy, unstructured data doesn't intimidate you — it interests you 1-3 years of experience; what matters more is what you've actually built
Why TitleWize
Early team, real ownership, hard problem. You'll have direct access to founders and work on decisions that shape the product for years. You will be a part of our interactions with our users, brainstrom with us, and ideate on scalable solutions.
Click on Apply to know more.