Use case · Forecasting

Autonomous AI forecasting — from CSV to deployed prediction API.

Drop your time-series data, describe the horizon, get a forecast. OctOpus picks the model family the data calls for — deep forecasting architectures, time-series foundation models, or tree-based models with lags — runs experiments, validates on a real out-of-sample holdout, and deploys a prediction endpoint.

TL;DR. Forecasting is OctOpus's strongest wedge. The agent moves between model families as the evidence warrants — tree-based models with lags and calendar features, modern deep forecasting architectures, time-series foundation models — keeping whichever wins on out-of-sample holdout error.

What OctOpus forecasts well

Models the agent rotates through

ApproachFamilyWhen the agent picks it
BaselineGradient boosting with lag, rolling, calendar featuresAlmost always — fast, strong, interpretable.
Tuned gradient boostingA different boosting family with automated hyperparameter searchWhen the baseline has headroom and the data justifies a search.
Deep / modernModern deep forecasting architecturesLong horizons, multi-series panels, exogenous signals.
FoundationTime-series foundation modelsZero-shot first; cold-start data; benchmark anchor.
StackingAn ensemble over diverse base learnersWhen the residuals of different families are uncorrelated.

How a forecasting run looks

  1. Profile. The agent reads the schema, detects the time column, infers granularity (daily / hourly / weekly), spots multiple series, and flags exogenous candidates.
  2. Plan. Writes a research spec: target, horizon, validation strategy (rolling-origin or fixed cutoff), and which families to try in which order.
  3. Run. Writes new training code for every experiment, executes it in a sandbox, reads the metrics.
  4. Diagnose. When something fails, the agent reads the traceback, writes a targeted fix (dtype, missing dependency, cardinality blow-up), and retries.
  5. Validate. The winner is scored on a holdout slice the LLM never sees.
  6. Deploy. A prediction endpoint plus a downloadable bundle with the full training code and the trained model.

What you get back

Forecast your data free → See benchmarks vs traditional AutoML