AnekantThe name is a working method.
"Anekant" comes from anekāntavāda — the Jain principle of many-sidedness. It says that no single viewpoint holds the whole truth; reality only comes into focus when you look at it from more than one angle.
That is exactly how we build software.
No single viewpoint holds the whole truth. So we examine every problem from the user's side, the data's side, and the model's side — before we write a line of code.
Deliberately small. On purpose.
Anekant AI is an AI development company in Ahmedabad, India, founded by Ketul Shah, an engineer. It is a small studio by design — not a stepping stone to an agency with account managers and handoff documents.
The trade we're making is deliberate: fewer projects, taken seriously. When you write to us, an engineer reads it. When we scope your project, the person scoping it is the person who will build it. When something breaks at a bad hour, you already have the builder's attention — there is no ticket queue between you and the fix.
We won't show you a team grid of stock-photo faces, because there isn't one. What you get instead is simpler: the person you talk to is the person who builds.
Ketul Shah
Engineer first, founder second. Builds across the stack — from data pipelines and LLM systems to the front ends that make them usable — and stays on every project from first call to production.
Based in Ahmedabad, Gujarat. Works with clients anywhere.
ketul.shah@anekantai.comFour principles we don't bend
Production over demos
A demo proves an idea; production proves a system. It isn't done until it runs — with monitoring, error handling, and someone accountable when it misbehaves. Every engagement ends with software in use, not a video of software.
Many-sided by default
Second opinions are built into the process — we argue with our own designs before you have to live with them. If a decision only makes sense from one angle, it isn't a decision yet; it's a bias with a deadline.
Plain answers
If AI is the wrong tool for your problem, we say so — even when it costs us the project. A cron job that works beats an agent that impresses. You'll always know what we think and why, in plain language.
Small and senior
No handoffs, no juniors learning on your budget. The engineer who understood your problem is the engineer who solves it. Small means less overhead in your invoice and no translation loss between "sales" and "delivery".
The stack we reach for
We're not dogmatic about tools — the problem picks the stack, not the other way around. But these are the ones we keep reaching for, because they've survived production more than once.
Tell us your side of the problem.
Write to us the way you'd explain it to a colleague. We'll look at it from a few more angles and reply within a day — with a plan, or with an honest "AI isn't the answer here".
Start the conversation