Applied AI, written for the people who have to decide.
Short, sourced pieces on the questions companies ask us before they build: which method, what it costs to run, where the risks are. Every piece is written by Vishal and states what we have and have not done ourselves.
25 Oct 2026
RAG vs fine-tuning: which one does your company actually need?
A decision table for assistants, classification, extraction and domain language, and the three myths that cost teams months.
25 Oct 2026
Synthetic data for Indian startups: when it helps, when it hurts, and the DPDP question
Small datasets, vendor sharing and testing; bias and false anonymity; and how synthetic data relates to the DPDP Act, 2023.
25 Oct 2026
Can a 20-person company run a private AI assistant on one machine?
What we learned running a 30-billion-parameter model with retrieval on a single workstation, and who it fits.