Shipped, not proposed.
We take on projects that put data and generative AI to work, mainly in healthcare and life sciences. Client names are withheld; each entry describes the problem and what we delivered.
Pharmaceutical company, data & analytics function — embedded support
Working inside the team on everything from day-to-day analysis to generative AI, continuously building several applications. We replaced functionality the team had been buying as an external service with in-house apps, and moved the aggregation and first-draft documents that staff repeated by hand onto generative AI. What they could only keep buying is now owned in-house, and the time lost to manual work goes back into the analysis itself.
- TypeScript
- React
- FastAPI
- Python
- R
Pharmaceutical company — a data governance foundation for RWD
Joined the build-out of a foundation for sharing and using real-world data across the organization. Beyond selecting and configuring the tooling, designed the operating rules and ownership model — who maintains what — and stayed through to the point where teams actually use it.
Academic research — analysis of large-scale medical data
Commissioned to run the analysis for a study using a large commercial health dataset and DPC (Japanese inpatient claims) data. Most of the effort went into turning the researchers’ “this kind of patient” into variable definitions that reproduce it — settling everything down to the observation windows across two datasets with different granularity and coding systems. The work carried through to an accepted paper.
- Python
- Alteryx
Healthcare data vendor — designing the offer to pharma
Clarified what pharmaceutical companies actually want, drawing both on time spent on the buying side and on fresh interviews. Rebuilt the offering and its targeting, the structure of proposals and sales materials, and a standard PoC format. Joined the client meetings as well, and the work led to a signed deal.
University spin-out biotech — business strategy as external CMO
For a fixed period, led business strategy as external CMO at a university spin-out developing early cancer detection technology: market sizing and competitive positioning, investor pitches and Q&A, and representing the company at academic congresses and in business meetings.
Major manufacturer — in-house data analysis training
Designed and delivered hands-on training that takes participants from acquiring public statistical open data through processing, visualization and analysis. No special tools — built so participants can reproduce it in their own work the same day. The client keeps coming back for further sessions.
AI SaaS company — business automation PoC
Joined a business automation PoC built on an AI workflow product as consultant and PM. Across several use cases, handled current-state analysis, to-be design, specifications for the engineers and client coordination — and designed and ran the evaluation of how far the AI’s output could be trusted.