NoiselessOS · Quantitative research platform
NoiselessOS is a personal research platform built around noiseless charts — Renko and Point & Figure. Chart engines, indicator and pattern libraries, and a multi-year backtesting suite, all in service of one goal: defined-risk, systematic income from index derivatives — tested before it is trusted, executed by rules rather than emotion.
01
Renko and P&F charts print only when price moves a fixed percentage — time disappears, structure remains, and signals cannot repaint. NoiselessOS turns that property into a research pipeline:
Renko and Point & Figure engines built from 1-minute data on an exact logarithmic grid — deterministic and repeatable: the same data always prints the same bricks.
Full libraries of trend-state indicators and price-structure patterns computed natively on bricks and columns — never borrowed from time charts.
Multi-year, multi-regime tests with costs and slippage modelled — and a hard rule that an edge which dies at realistic slippage was never an edge.
Strategies that pass the bar run as rules — defined-risk credit-spread structures, entries and exits taken by algorithm, worst case written down before entry.
02
The goal is not to trade more. The goal is to not trade unless the market pays for it — and to know the worst case before the first rupee moves.
03
NoiselessOS is being built toward a properly regulated future, in a deliberate order:
Nothing is on offer today — no signals, no subscriptions, no managed money. If you would like to hear when that changes, write "keep me posted" to [email protected].
04
Dushyant Thakker · CFA · FRM · About 20 years in various roles at global banks and asset managers, most recently at JP Morgan Chase Asset Management. NoiselessOS started as the research stack behind my own systematic, defined-risk index strategies, tested like it's someone else's money. It is now a research hub that is developed and maintained every week, and the studies on dushyant.co.uk are run on it.
How it is built. The ideas, the research questions, the direction and every decision about what gets tested, kept or dropped are mine. The architect and developer is Claude, the AI model made by Anthropic. Claude designs and writes the code, runs the tests and drafts the research write-ups. I decide what is published.