AI & Data Expert - Flexible Day Rate Engagement Remote | Day Rate Open | Flexible Commitment
I work exclusively with mission-led tech companies, and one of my favourite clients needs some urgent AI & Data support. They are a fast-growing fintech on a mission that genuinely matters and the work you'd be doing is high-impact from day one.
The opportunity
This isn't a standard role and we're not looking for a standard candidate. The engagement is fully flexible it could be a block of days, an ongoing advisory arrangement, or somewhere in between. We'll shape it around the right person.
What you'll be working on
You'll be embedded as a trusted expert across five core challenge areas:
* RAG & PDF extraction - building and improving retrieval-augmented generation pipelines that actually work with real-world financial documents. Not clean text. The messy, table-heavy, footnote-riddled kind.
* Data validation at scale - designing reliable approaches that combine AI automation with human review, because you know AI gets it wrong and you build the safety nets accordingly
* Financial & governance data management - extracting, structuring and maintaining large, complex datasets in a high-stakes environment
* Prompt engineering - systematic, evidence-based design and testing of prompts for data extraction and classification. Not vibes. Not iteration for its own sake.
* Model strategy - advising on LLM selection, cost/accuracy/latency trade-offs, and managing model updates over time. When someone asks which model to use and why, you have a real answer.
Who we're looking for
You've been in the trenches. You've shipped real systems, dealt with genuinely messy data, and have strong opinions formed from production experience.
Specifically, you'll have:
* Proven, hands-on RAG experience with unstructured documents - PDFs, filings, reports
* Built validation pipelines that combine automated checks with human-in-the-loop review
* Made deliberate, documented LLM selection decisions and lived with the consequences
* Strong, systematic prompt engineering skills
* Worked across multiple AI models and navigated updates and changes in a live environment
* A background in fintech, ESG, governance, legal tech or any domain where data accuracy carries real consequence - desirable but not essential
The one question I'll ask you
Tell me about a time AI got the data wrong and you didn't catch it immediately - what happened and what did you change?
If you have a great answer to that, I want to hear from you.
The detail
* Fully remote or London or Bristol
* Day rate open to discussion - Outside IR35
* Flexible commitment - advisory, a block of days, or ongoing retainer
* Start: as soon as possible