US and India, one terminal
State the mandate. Every strategy engine proves itself against its benchmark before anything funds it. The winner runs forward on live prices with a desk of AI analysts on it, a bull and a bear arguing every name, and a grade on the book from Sarathi. Every trade waits for you.
Research tooling, not investment advice. Backtests are simulations; nothing here is a recommendation.
25 of 29 scored holdings are in the top half of the universe
vol 14.2%, balanced band 5%-20%
EA: not scored, 7.6% of the book unjudged
top-5 = 38%, HHI 0.048
Mandate has no sector policy to drift from.
0.0%, at policy
Nothing. The book sits inside its mandate today, so no trade is proposed. When one is, it arrives with the reason and waits for you.
The platform’s own review, retrieved on 23 Sept 2026, 08:46 UTC. Sign in and the same desk grades your book.
Strategy Lab
Two demo mandates, one on each market, swept under every engine over the same history. The lab exists to say which engines beat the index on that mandate and which did not, before a rupee or a dollar moves. What it found is below, as it is stored.
| Engine | Rank | CAGR | Sharpe | Max DD | vs index |
|---|---|---|---|---|---|
| Sector Rotation | 1 | 19.3% | 1.08 | -11.8% | beat |
| Trend Regime | 2 | 15.1% | 0.93 | -12.0% | beat |
| Engine 3 | 3 | 15.1% | 0.92 | -13.0% | beat |
| Engine 4 | 4 | 15.1% | 0.92 | -13.0% | beat |
| Engine 5 | 5 | 13.4% | 0.85 | -13.0% | lagged |
| Engine 6 | 6 | 11.8% | 0.84 | -12.9% | lagged |
| Engine 7 | 7 | 11.0% | 0.76 | -13.9% | lagged |
| Engine 8 | 8 | 10.3% | 0.71 | -15.0% | lagged |
| Engine 9 | 9 | 10.3% | 0.71 | -15.0% | lagged |
| Engine 10 | 10 | 8.6% | 0.65 | -11.1% | lagged |
| Engine 11 | 11 | 6.2% | 0.63 | -11.7% | lagged |
| Engine 12 | 12 | 3.0% | 0.35 | -11.7% | lagged |
| Engine | Rank | CAGR | Sharpe | Max DD | vs index |
|---|---|---|---|---|---|
| India Volume Reversal | 1 | 16.2% | 1.43 | -9.0% | beat |
| 52‑Week High | 2 | 15.0% | 1.42 | -7.9% | beat |
| Engine 3 | 3 | 11.7% | 1.33 | -9.9% | beat |
| Engine 4 | 4 | 16.0% | 1.29 | -9.9% | beat |
| Engine 5 | 5 | 14.3% | 1.25 | -10.1% | beat |
| Engine 6 | 6 | 14.3% | 1.25 | -10.1% | beat |
| Engine 7 | 7 | 13.8% | 1.25 | -9.9% | beat |
| Engine 8 | 8 | 13.4% | 1.22 | -10.4% | beat |
| Engine 9 | 9 | 13.4% | 1.14 | -12.1% | beat |
| Engine 10 | 10 | 14.6% | 1.11 | -13.0% | beat |
| Engine 11 | 11 | 14.6% | 1.11 | -13.0% | beat |
| Engine 12 | 12 | 12.7% | 1.05 | -16.1% | beat |
Every curve and figure above is a stored run, retrieved from the platform on 23 Sept 2026, 08:46 UTC. Engines whose curves coincide on a mandate are shown as the lab stores them.
A mandate is a statement of what a book should be: universe, sectors, risk profile, how many names, cash reserve, stop-loss. Draft it yourself or describe it in plain words and let the AI draft it.
Every engine runs it over history, benchmark-relative, and the lab ranks them on risk-adjusted return. Nothing funds a book on a hunch.
The winner runs forward as a paper book on live prices. Rebalances arrive on your schedule with turnover, sector drift and a written reason per name. You accept or reject.
A desk of analysts, a bull and a bear on each holding, the Telegram buzz on your names, and a Sarathi grade on the whole book, every session.
What it holds
From a mandate to a book that runs.
Universe, sectors, risk profile, position count and cash reserve, with a composition panel showing what you can actually buy.
One mandate under every engine at once, ranked on risk-adjusted return, curves stored.
Backtest your own stock list, save it, run it as a book.
Cash or funded paper books on live prices, in USD or INR, with rebalances proposed and explained on a schedule you set.
A grade on your book, what it would change and why, and a written letter you can act on.
Read a Zerodha or SnapTrade account beside your paper books. Nothing is placed for you.
The evidence under every decision.
A 24-factor read per name, its ownership (13F, fund holdings, mutual-fund schemes, promoter pledge and SAST), filings and a decision snapshot.
What-if from revenue through margin to EPS and multiple.
Who reports next, our forward EPS beside the street's, call sentiment, an AI read of the 8-K, and post-earnings drift.
An event in one company and the names it reaches, with the business linkage, not just the correlation.
Monte Carlo across size and sector: how many names actually diversify.
The strategy's target book and its stored, benchmark-relative history.
The session, as it happens.
The session that just closed: movers, breadth, earnings, news, and what wants a decision from you.
Your book and the day's exceptions on one screen.
The realtime engine and its signals during market hours.
Market news mapped to tickers and scored, for the US and India.
What Telegram channels are saying, by ticker, with each call judged at the close and each channel graded.
S&P 500 to S&P 600, Dow 30, Nifty 50 to the total market, and sector slices of both.
Grounded in what the platform computed.
Opposing cases argued per security, then judged, so the weak side of a thesis is visible.
Agents on regime, macro, risk, concentration and news, all writing from one fact pack.
Ask about exposure, risk or a holding; the answer is computed from your positions.
Every screen answers questions about what it is showing, from the numbers on it.
Underneath
10-K, 10-Q and 8-K filings, XBRL facts, earnings releases, Forms 3/4/5, 13F, 13D/G and N-PORT fund holdings
Bhavcopy prices, corporate filings, Reg-31 shareholding, pledge and SAST disclosures, mutual-fund scheme portfolios
Daily bars for both markets, corporate-action adjusted, index membership, with a data-quality verdict per security
FRED macro panel, news mapped to tickers and scored, and Telegram channels read and graded on their calls
Counts as the platform reported them on 23 Sept 2026, 08:46 UTC.
Signing in with Google opens a ten day trial of the full terminal.
Open the terminal