Max Lang Oxford researcher header image

stay hungry, stay foolish

Max Melchior Lang

About

I am a DPhil Student at the University of Oxford and passionate about answering the question: How do we gather high-quality data and extract valuable information to answer important questions?

You can contact me here.

More info

I like to think of statistics as the honest interpretation of data, even when the honest interpretation is less catchy than the story we might want to tell. Whether we're fitting models to reality or trying to fit reality to our models is worth thinking about. The latter defeats the point, though it is often more marketable.

At Oxford, my work focuses on modeling the co-distribution of Malaria and Schistosomiasis. This involves a lot of time thinking about generalized additive "smooth" models and analyzing satellite imagery in various geometric configurations. I'm part of the SchistoTrack cohort, supervised by Goylette Chami and Christl Donnelly.

Previously, under the supervision of Frauke Kreuter in Munich, I worked on the rather unfashionable problem of simply asking good questions. Understanding that people would rather click boxes than type thoughtful responses often matters more than any sophisticated model.

This led to several engineering projects to modernize data collection:

I'm always happy to discuss research ideas and look forward to new collaborations.

Publications

2025

2024

2023

Talks

2025

2024

2023

Projects

Ongoing

  • Malaria-Schistosomiasis co-distribution – Modeling disease co-infections in rural Uganda using satellite imagery and spatial statistics. DPhil project with SchistoTrack Group.
  • AVA – AI vaccination assistant for patients over 60 with mild cognitive impairment. Lead developer of the voice agent architecture. Registered RCT, BMBF-funded. [LMU study page] [Trial registry]
  • AI Telephone Surveys – Automating telephone surveys with conversational AI for 60 Decibels. Lead developer of the end-to-end architecture. 20,000+ interviews and counting, now scaling in emerging markets. [Paper] [Case study]

Open Source

  • hexsmoothR – R package for hexagonal grid smoothing of satellite data
  • Pytector – LLM prompt injection detection Python package
  • RAG-nificent – Chainlit RAG application with Pinecone pipeline
  • cuda-gis-smoothing – CUDA-accelerated GIS smoothing algorithms
  • timeglobe – Like worldtimebuddy.com but awesome

See GitHub for all projects →

Awards & Funding

My research is graciously funded by Oxford Population Health.

Teaching & Writing

Contact

Max Melchior Lang

Big Data Institute
Old Road Campus, University of Oxford
Oxford OX3 7LF, United Kingdom

Email: max.lang[at]stx.ox.ac.uk

ORCID: 0009-0004-6815-5321