/jobs/ai-engineer
AI engineer
agent quality. prompt engineering, LLM evaluation, memory retrieval, tool design.
remote10-20 hrs/weekcontributorEUR 80-130/hr
about doobls
doobls turns human expertise into AI agents you own. 3-layer memory, multi-model orchestration, encrypted multi-tenant architecture. open source core. EU-sovereign. pre-series A with paying customers.
the work
- 3-layer memory: episodic events, narrative clustering, strategic workflows. make retrieval fast and accurate.
- evaluation frameworks. automated evals, regression detection, quality scoring across models.
- prompt architecture for multi-step agent workflows. tool selection, context management.
- integrate and evaluate new foundation models as they ship.
you
- shipped LLM-powered products in production
- prompting, function calling, multi-step agent architectures
- model quality evaluation. metrics over vibes.
- strong python. comfortable turning papers into code.
compensation
- EUR 100/hr cash or EUR 150/hr equivalent in BKY (tokenized equity). you choose your mix.
hiring process
- 1. answer the questions below
- 2. conversation (45 min)
- 3. paid trial (1-2 weeks)
- 4. offer
apply
send your answers to jobs@baikery.studio
- 1. where do current agent frameworks fail on memory and context?
- 2. an LLM system you built where reliability was the hard part. what did you learn?
the future of work is changing with AI, and you get to shape what that looks like for you.