← All portfolios

01Portfolio · RAG & enterprise search

Retrieval systems that survive real documents

Production RAG is most of my working week: pipelines, hybrid retrieval, grounded citations, and the hardening that keeps a live system safe. These are real contracts, figures as recorded on Upwork.

  • 2,600+hours on RAG / LLM work
  • $85K+billed across RAG contracts
  • Activeflagship still running

01Delivered work

Projects on the record

Client names withheld under contract — every figure is as recorded on the Upwork profile.

ActiveEnterprise document intelligence

Production RAG + Azure OpenAI platform

The flagship: a live retrieval system for one client across two contracts and a year of continuous work. Rebuilt ingestion around structure-aware chunking, moved retrieval to hybrid vector + BM25 with reranking, added a citation contract — then hardened the live system against prompt injection and data leakage once real users arrived.

  • 1,970hours, both phases
  • $80.6Kbilled to date
  • 4.9★phase-one rating
  • Azure OpenAI
  • LangChain
  • Vector DB
  • Hybrid retrieval
  • Injection defence
Legal tech

“Court-ready” accuracy for a legal search engine

Senior RAG engineering on a legal search product: built a legal taxonomy over the corpus and pushed answer accuracy to a bar where results could be relied on in practice — with documentation the client called out by name.

  • 5.0★contract rating
  • Fixedscope & price
“Excellent work, communication and documentation. They delivered a working product exactly as I asked for — an A+ experience all-around. I will be hiring Wahab again.”
  • Legal taxonomy
  • Eval harness
  • Citations
Software delivery

LLM integration for a product team

A short-term senior engagement embedding LLM features into a client's product — prompt architecture, retrieval wiring and delivery support through launch.

  • $2.5Kfixed engagement
  • 5.0★contract rating
“Wahab has been extremely easy to work with, and very professional. His expertise has been invaluable in helping us to deliver this project.”
  • LLM integration
  • Prompt architecture
  • Delivery

02Systems of this class

Recognise the shape

Well-known products of the same class — unaffiliated reference points, so a first conversation starts from a shape you already know.

Need RAG & enterprise search work like this?

Send the problem and where it's stuck. I'll tell you honestly whether I'm the right person and what it would take — usually within one business day.