RAG on your own documents
Connect the documents your team already uses. Preserve permissions, retrieve the right evidence and make every answer traceable to its source.
AI ENGINEERING
AI your team can put to work.
I'm Adam Boudjemaa, an AI engineer in France. I build production agents and RAG systems, with evaluations, permissions and a handover your team can use.
France · French and English · Freelance or permanent roles
Where I can help
Agents for real workflows
Build bounded tools and approval steps around reconciliation, reporting or operations. Define what the agent can do and when a person takes over.
Evaluations before rollout
Test missing evidence, wrong numbers and tenant boundaries. Add abstention, cost limits and regression checks before giving a workflow more autonomy.
The work behind the profile
A production platform, built inside the client
I delivered an 11-agent platform for a confidential US real-estate fund. The work included FastAPI, Postgres/pgvector, document ingestion, verified citations and numeric checks. Eleven describes the initial delivery.
Read the delivery case studyEngineering ownership through operations
As CTO at Integra, I built the engineering team and led infrastructure through launch preparation. That operating background shapes how I handle production AI: permissions, failures and handover belong in the design.
Review my experience and CVHow we start
- Pick one workflow and define a measurable acceptance test.
- Inspect data access, document quality and the current system.
- Ship a narrow end-to-end path with evaluations and human controls.
- Hand over code, operating notes and the evidence needed to extend it.
Before you get in touch
Can I hire you for a permanent role or a freelance mission?
Yes. I'm open to permanent AI engineering roles and scoped freelance missions. Send the role or business problem, your stack, timeline and budget or salary range so we can assess fit.
Where are you based, and do you work remotely?
I'm based in France, an EU citizen, and work in French and English. I'm open to remote work, on-site engagements and relocation, depending on the role.
Which AI systems have you actually built?
I delivered a production multi-agent platform for a confidential US real-estate fund, including RAG on FastAPI and Postgres/pgvector, Amazon Bedrock, document extraction, verified citations, numeric checks and explicit abstention. The public case study describes the initial 11-agent delivery without naming the client.
Can you improve an existing prototype?
Yes. A useful starting scope is to inspect one workflow, its source data and failure cases, then agree on a release gate. That may lead to a retrieval fix, a smaller agent design or stronger evaluations before adding features.