Ankit Choudhary @choudharyankit
AI Architect · Agentic AI · AI Systems · IIT Kanpur
I have been building data and AI systems for more than eight years, from search and NLP to
generative AI and agentic systems. I write about what it takes to make them work outside a
demo, and about what economics, political campaigns, and teaching taught me along the way.
Writing OCT 28 History of NLP: From TF-IDF to BERT, LLMs and AI Agents, a Practitioner's View How NLP and AI changed from statistics and random forests to BERT, RAG and agents, as one practitioner saw it, and why I now care about meaning. OCT 19 What Is Ontology in AI? Giving AI Agents Business Context What is ontology, where can a business use it, and how does it help AI? A practical look at shared meanings, connected information and business decisions. OCT 14 If Agent Skills Are Easy to Copy, Where Is the Moat? Skills and agent workflows are useful, but are they a business moat? I think the stronger opportunity is a harness that understands how a specific domain works. OCT 14 Where Could Specialised Models Fit in an Agent Workflow? Smaller, specialised AI models may improve specific steps in an agent workflow. Here is how I would evaluate their accuracy, cost, integration and long-term maintenance. OCT 9 AI makes IT services more productive. Will it make them more profitable? How generative AI changes delivery costs, pricing and value capture in IT services OCT 3 System One Models: Jev vs Laya and How They Change AI Agents System One models return fast, structured decisions instead of prose. Learn how Jev and Laya compare and where they fit in AI agents. SEP 14 What an AI-Ready/AI-Native Workforce Actually Looks Like Why building an AI-ready workforce starts with redesigning workflows, not training people on tools, and what it means for engineers, managers, juniors and learning. JUL 16 AI Agents as Users: What Changes When Software's Customers Aren't People (Part 2) If AI agents become software's primary users, interfaces change: predictable outputs, clear boundaries and identity matter more than screens. Part 2 of a series. JUL 15 Software 3.0 Explained: Why We Need to Rethink User Interfaces (Part 1) Software 3.0 shifts users from running workflows to stating intent. Why that changes how we think about interfaces, from cab booking to prior authorization. Part 1. JUL 8 Agentic AI for Banking: A Governance Checklist What banks need to check before using autonomous agents in software delivery: approvals, independent reviews, audit trails, data controls and accountability. JUL 7 Not Every AI Engineering Company Is Solving the Same Problem Buyers compare AI engineering partners by size, but what a company is optimised for matters more. A way to tell IT services, digital engineering and AI-native apart. FEB 12 Do We Really Need Knowledge Graphs in the Age of LLMs? For most problems, I think knowledge graphs are over-engineering. A case for multi-agent systems with chain-of-thought reasoning as the more adaptable alternative. FEB 10 Why Agentic AI is the Missing Link for Real-World Applications of LLMs LLMs have mostly assisted people rather than worked on their own. Why agentic AI, with validation and audit trails, is the structure that closes that gap. FEB 9 Reforming the Indian Judiciary: Can Technology Solve the Backlog Crisis? Where AI, connected court records and legal technology could help India's courts, and why changes to governance and legal procedures still matter. FEB 7 Agentic AI: Architecting Autonomous Systems for Complex Workflows An introduction to agentic AI architecture: a workflow of specialised nodes with a manager, shared memory and a human feedback loop, rather than one monolithic model. NOV 1 Optimizing AI Models with Quantization: Enhancing Efficiency in Resource-Constrained Environments How quantization shrinks AI models and speeds up inference for resource-constrained environments, and the formats and methods behind it, from GGML to GPTQ. OCT 30 Fine-Tuning vs RAG: Choosing the Right Approach for Your Question-Answering System Fine-tuning or RAG for a question-answering system? Eight factors I use to decide, from task goals and latency to how often your information changes. OCT 29 RAG Systems: An Innovative Approach to Question-Answering RAG tackles two limits of LLMs, hallucination and outdated knowledge, without retraining. How retrieval and generation fit together and the design choices involved.