The Forecasting Company GTM brain

The Forecasting Company offers time-series foundation models and forecasting tools for data and planning teams in retail, energy and other industries.

category

Time-series foundation models and forecasting

market profile

  • text: Organizations forecast many time series across demand, energy load, prices and workforce planning, often with diverse data and horizons.
  • text: The company offers open-weight t0-gamma, an API and a backtesting interface, combining a research-led model with practical evaluation tools.

personas

  • title: Demand planning director
    pains:
    • Forecasts across many SKUs are difficult to backtest consistently
    • Poor demand estimates complicate inventory and procurement decisions
  • title: Forecasting or data science lead
    pains:
    • Maintaining domain-specific models and reshaping data take engineering time
    • Evaluation across all series can be computationally expensive
  • title: Energy forecasting manager
    pains:
    • Grid load and price predictions need to reflect changing historical patterns
    • Planning decisions require usable uncertainty ranges

icp

  • text: Retail and consumer goods organizations with many SKU-level demand series and recurring inventory decisions; firm size is not stated.
  • text: Energy operators and other organizations with high-volume load, price or workforce time-series forecasting needs.
  • text: Technical teams that can evaluate open weights, a forecasting API or backtesting against existing data.

message pillars

  • text: Forecast different domains using one model without per-domain training, according to the company.
  • text: Describe the data once and generate calibrated forecast distributions without feature engineering or fine-tuning, per the site.
  • text: Backtest broadly against historical actuals and judge performance on the buyer's own data.
  • text: Use open-weight t0-gamma and explore the API or Retrocast interface.

objections

  • objection: Will a general model beat our existing forecasts?
    response: Backtest against current baselines across representative series, horizons and business metrics before deployment.
  • objection: Can the model handle our particular operating context?
    response: Test with actual history and relevant context, including known future variables where supported, and inspect prediction intervals.

competitors

  • name: Nixtla
    website: https://www.gartner.com/reviews/product/nixtla-916571571/alternatives
    why: TimeGPT and forecasting libraries address time-series prediction for technical teams.
  • name: Lokad
    website: https://www.lokad.com/demand-forecasting/
    why: Demand forecasting and supply-chain optimization compete for planning use cases.
  • name: o9 Solutions
    website: https://www.gartner.com/reviews/product/o9-solutions-demand-planning/alternatives
    why: Demand planning software is a broader enterprise alternative for the same planning budgets.

influencers

  • name: Geoffrey Négiar
    url: https://www.linkedin.com/posts/geoffrey-negiar_launch-yc-the-forecasting-company-foundation-activity-7224916536078053377-JvhK
    why: Co-founder and CEO who publicly announced the company and has forecasting research experience.
  • name: Joachim Fainberg
    url: https://www.linkedin.com/posts/jfainberg_launch-yc-the-forecasting-company-foundation-activity-7224920704255688704-Oqjq
    why: Co-founder and CTO who shared the YC launch.

keywords

  • time series foundation model
  • probabilistic demand forecasting
  • SKU demand forecasting
  • energy load forecasting AI
  • forecast backtesting software
  • zero shot time series forecasting
  • forecasting API

voice

  • Research-led and precise
  • Optimistic about general-purpose models
  • Minimal and visual

signals

  • text: A planning team needs to forecast a large catalog of SKU series and evaluate historical errors at scale.
  • text: An energy team is evaluating load or price forecasts for operational planning.
  • text: An ML team is evaluating open-weight time-series models or replacing per-series models.

stack

  • forecasting API
  • Hugging Face
  • SQL

pricing

model: unknown

motion

hybrid

channels

  • LinkedIn
  • Hugging Face
  • research content
  • developer documentation
  • YC directory

confidence

medium

Founders

  • name: Geoffrey Negiar
    title: Founder/CEO
    role: CEO
    bio: Co-founder and CEO at The Forecasting Company, helping enterprise customers improve their operational decision making. PhD grad from UC Berkeley's BAIR. I spent time with Amazon Forecasting Science in SCOT, Google Brain, and Bloomberg LP where I worked on forecasting, optimization and deep learning research.
  • name: Joachim Fainberg
    title: Founder/CTO
    role: CTO
    bio: PhD in ML for speech and full-stack developer. Interned at Sonos where I built an on-device wake word detector and at Bloomberg building a large ASR system. Worked full-time at JPMorgan in their Machine Learning Center of Excellence in New York, and then at two startups, team lead ML/NLP at Consigli, a prop-tech startup, and sr algorithm engineer and full-stack at Vind AI, a SaaS for designing off-shore wind parks.

Back to YC companies