I think in public.
100+ publications as DrSwarnenduAI · 13K+ followers · Updated every week.
I write about what actually breaks when you move AI from a notebook into a company — agentic systems, LLMOps, forecasting and the governance that keeps them honest. The aim is to explain what works in practice, and what is theatre dressed as transformation.
Featured writing
A few pieces that show how I reason. Each links straight through.
GPT-4 Is Just a Giant Markov Chain — And That's the Genius of It
Why statistical structure, not "intelligence", explains a surprising amount of what large models can do — and what they can't.
Gradient BoostingWhy CatBoost Mathematically Outsmarts XGBoost
A first-principles tour of boosting — ordered target statistics, and when LightGBM, XGBoost and CatBoost each win.
Forecasting · ProductionStop Forcing Your Time-Series Models to Memorize the Universe
Why zero-shot, retrieval-grounded forecasting beats memorising every series — the thinking behind RAFT-style architectures.
Foundations · Running thesis · GitHub ★22AI, Mathematically Explained
The running thesis: modern AI is classical mathematics with better computers. Repo + essays — the foundation of everything I build.
What I write about
Agentic AI & LLMs
Multi-agent orchestration, LangGraph state machines, RAG pipelines, MCP, production failure modes and how to build systems that don't hallucinate money.
Forecasting & optimization
Time-series architectures, Temporal Fusion Transformers, zero-shot forecasting, Bayesian methods, Monte Carlo simulation — production not research.
The mathematics of AI
Why modern deep learning is classical mathematics with better hardware — gradient flows, matrix factorisation, attention as soft routing, all from first principles.
Speaking & workshops
Keynote at the International Forecasting Forum and MLDS on predictive orchestration, forecasting architectures and responsible agentic AI in the enterprise.
AIM 40 Under 40 · 2026
Named a Top AI Innovator by Analytics India Magazine for enterprise AI impact at scale.
13K+ followers
Readers across Medium, Substack and Towards AI following weekly writing on production AI and forecasting.
Keynotes & summits
Machine Learning Developers Summit (MLDS) and the International Forecasting Forum. Available for keynote, panel and workshop.
What people I've worked with say
"Swarnendu is made for analytics. He is humble, eager to learn and open to listening, even though he already has very in-depth knowledge of the field. On top of all that, he is hard-working and very good at Python coding. He is a great asset to any analytics team."
Suvendu Tripathy — Associate Partner, Cognizant · Aug 2022
"I had the pleasure of working with Swarnendu during his time at Cognizant. I was constantly amazed not only by his technical skills, but by how well he communicated them to business users, and his ability to connect the dots to surface real insights. I couldn't recommend the experience more."
Mishtert T — Enterprise AI Leader · Senior colleague, Cognizant · Jul 2022