Quant Lab
Research notebooks with the datasets loaded, model training on our compute, backtests over real history and tearsheets you can share.
Quant Lab is Sigmo's workspace for data-driven research: a place to dig into the numbers behind a market, test an idea properly and build a repeatable process instead of relying on gut feel. It is the part of Sigmo that a hedge fund would recognise.
Notebooks with the data already there
Open a hosted Python notebook and the datasets are loaded: price history at every timeframe, fundamentals, filings and events, options chains and snapshots, alternative data. The Sigmo SDK gives you one call to inspect coverage, peek at a dataset's shape and pull a bounded range. Point-in-time panels keep you honest: a period's date is not its publication date, and the readers respect that.
Ask, or code
Tell the assistant what you want to test and it writes and runs the notebook, or open the notebook and write it yourself. Either way the figures, tables and code stay together, so a result can be inspected and reproduced later.
Train models on our compute
Train market-level machine-learning models on a universe of symbols with a chosen feature set and labelling scheme, on Sigmo's compute rather than your laptop. Runs are queued, tracked and reproducible; each model records its configuration, its data window and its metrics. ML Model Training has the full picture.
Backtest over real history
Run any model or rule-based strategy over real historical data with the portfolio construction you choose, and get returns, drawdowns, the trade list and a tearsheet. Treat the result as evidence about the test under its assumptions, not as a forecast; Sigmo shows you the assumptions so you can argue with them.
Regimes, correlations, seasonality, drawdowns
Test a thesis across market regimes. Study correlations and drawdowns, uncover seasonality, and see how similar setups behaved historically, across hundreds of datasets that are cleaned, aligned and connected.
Rankings, factors and strategy trees
Rank a universe on the factors you care about, combine signals into a strategy tree you can read, and turn a finding into a strategy the agent can run on paper.
Included compute
Pro plans include research notebooks and daily model-training runs. Max adds included compute credits every month for heavier research. The current allowances are on the pricing page.
See it on your own watchlist
Sigmo is free to start. Open the app, ask one question about a ticker you follow, and take it from there.