Max Held
Freelance developer · Data products & infrastructure
I build data products from end to end: from the underlying data to the app people use and the infrastructure it runs on.
Translated: “To Unix, through hardships”.
The Unix philosophy, summarised:
Write programs that do one thing and do it well. Write programs to work together. Write programs to handle text streams, because that is a universal interface.
12 years of experience in industry, research and open source. Clients include Posit (RStudio), Immunai and companies in finance, insurance and automotive.
Services
Declarative DevOps
Builds that only work on one machine? Deployments that need a checklist?
I turn environments, CI/CD and cloud resources into versioned configuration. Every build is reproducible and every change is reviewed.
Functional Data Products
A model stuck in a spreadsheet? A Shiny app that has become slow and fragile?
I build R packages, pipelines and Shiny apps from small, tested functions over tidy data. They are easy to verify, and easy to change.
Self-Sovereign(ish) Compute
Want to run AI and other workloads without handing your data to someone else?
I set up self-hosted and local-first systems on hardware you control. Where a managed service is the better choice, I will tell you.
Experience
Selected client projects, research and open source work, by industry.
I implemented a default risk model as a Shiny app for a national financial institution, and rebuilt a spreadsheet-based financial model in R, with a Shiny frontend and an API, cross-validated against the legacy Excel results. I also improved the performance and maintainability of a large Shiny app used by hundreds of internal users every week, and consulted on containerisation and cloud migration to Azure.
I built several Shiny apps for internal use at an international insurance group, from requirements engineering to deployment and performance tuning.
For Immunai, I standardised the compute environment for research and production workloads, across several locations, with dozens of internal modules and hundreds of external dependencies. Images were fully versioned, efficiently cached and rebuilt on every commit. I also productionised several internal R packages.
I consulted on inventory management automation in R, reviewed a large Shiny app for an automotive company, and worked with industrial protocols such as Modbus and EIB/KNX.
I worked as a researcher on opinion dynamics and collective decisions, including behavioural experiments and agent-based models. I wrote open source software for Q methodology, a psychometric method, and built ETL pipelines and dashboards for bibliometric data at a university library.
Let’s talk.
Tell me what you are building, where you are stuck, and your timeline.
I will get back to you within a few days.