About
Who builds Continuous Function, and why.
I’m Archit Khare, an applied researcher and machine-learning engineer. Continuous Function is my independent project: interactive notebooks and investigations that explain how modern machine learning works, a reader for paired evaluation scores under declared assumptions, and research proposals written before any study is run.
See selected workBackground
I have worked for more than seven years across deep learning, foundation models, language systems, information retrieval, fine-tuning, recommendation systems and large-scale experimentation.
Continuous Function has been my full-time focus since June 2026.
What I am building
Continuous Function has four parts. Each is usable on its own; they do not yet form one connected research system.
Why
I want people to keep the understanding and the judgment to decide which questions matter, what would count as evidence, and whether a result actually supports a claim.
The learning pages present machine-learning mechanisms through explanations and calculations, and Verify makes the assumptions used to read paired scores explicit. The longer-range aim, a place where people and AI agents investigate questions together while people keep authority over the questions and the evidence, is a direction, not something the site does today.
Status
The site has no accounts, advertising or analytics, and its tools run in your browser. The learning pages have not been evaluated in a study with learners. The research pages are proposals: no study has been run and no results are reported. The source repository is private.
Contact
Email founder@continuousfunction.ai. I am also on LinkedIn and GitHub. For a short tour of what to inspect first, see selected work.