About · Software engineer & systematic market researcher
A developer who tests every market idea in code.
I'm Amar, a software engineer in India building quantitative research and order-flow tools. DeltaStack is my engineering lab: I write code to parse tick data, rebuild candles, and test market hypotheses on Indian indices — then publish the method and every result, especially the ones that fail.
“If a number would not have been known at the moment of entry, it does not count.”
- Order flow first: who crossed the spread, and whether price agreed.
- Every test written down before it runs, with its pass bar.
- Tuned on one year, judged on another, after real costs.
- Only closed candles. No look-ahead, ever.
- Paper on unseen days before a single rupee is risked.
Why publish failures?
Because that's where most of the learning is.
honesty
A ledger with only wins is a sales page. Showing what failed is how you know the passes mean something.
usefulness
If a popular setup failed on our data, you can skip months of trying it yourself.
discipline
Writing the result down in public makes it impossible to quietly re-tune a losing idea.
limits
This is research and education. No calls, no tips, no managed money — and no promises.