Data tool · Web app
RI2 — Rapid Insights Data Engine
A no-code workspace for cleaning, visualizing and statistically testing tabular data, with real pandas and scikit-learn running in the browser through WebAssembly.
Data scientist · AI/ML engineer Massachusetts, USA
Hi, I’m Sudhanshu.
I build intelligent things.
সুধাংশু মুখার্জি
I’m a data scientist and AI/ML engineer building intelligent applications, language models, and practical developer tools.
01Selected work
Things I’ve built recently
Language models · Research
TinyLM Lab
Pretraining a 26.7M-parameter Llama-style model from random initialization on one laptop GPU, then checking what it learned against memorization, leakage and controlled ablations.
Web app · Product design
Kaku — Tiny Editor
A calm, private writing room in a browser tab: rich text, stationery-inspired themes and local export, with nothing ever saved or uploaded.
Education · Analytics
The Power BI Fellowship
A free, open-source way to learn Power BI by doing the job: checkable skill exercises, then tickets, incidents and architecture decisions at a fictional company.
Generative AI · Document intelligence
Self-Correcting Multi-Document RAG
Question answering and summarization over multiple documents, with agents that filter irrelevant context, score each answer’s factual consistency and regenerate weak answers.
02Focus
Three threads run through my work
-
AI & ML engineering
Training and evaluating models carefully: a small language model from scratch, retrieval that checks its own answers, multimodal fine-tuning and architecture benchmarks.
See
-
Developer tools & products
Software that removes a step from someone’s day: a data engine that runs in the browser, a terminal data toolkit and a private writing room.
See
-
Data, analytics & teaching
Years of BI work, forecasting and sports analytics, and passing it on through workshops and open curricula.
See
04Writing
Notes from the Hustle Data newsletter
-
20 SQL Commands For Your Next Data Science Job
The SQL you are most likely to be asked about in a data science interview, one command at a time.
-
How to Evaluate Machine Learning Models using Classification Metrics: ROC, AUC, Precision, Recall, and Beyond
What each classification metric actually measures, with reusable code for every one of them.
-
Deep Learning is all about Gradient Descent Algorithm
Backpropagation, gradient descent and stochastic gradient descent, and how they drive learning in neural networks.