Foundation models for time series
We are building the forecasting foundation model to rule them all. All enterprise companies run forecasting to plan their operations: staffing, supply-chain management, finances… We provide the data, models and platform to easily build the most accurate forecasts. We significantly reduce waste and increase cash flow for our customers.
Data is the lifeblood of our product. As the founding data engineer, you will build solid data pipelines for two main use cases: massive datasets for model training, and streaming data for forecasting.
You love your craft, have high standards, stay up-to-date with the latest tech, and know when to make trade-offs to ship.
Design scalable architectures capable of handling large, diverse datasets for model training, integrating multimodal inputs like numerical time series, text, and image data.
Develop real-time ingestion systems to ensure up-to-date data availability for operational forecasting, optimizing latency and reliability.
Build indexing, search and examination pipelines to explore the data and recommend relevant data sources for a specific use-case.
Define robust and extensible schemas to support both structured (e.g., CSV, databases) and unstructured data (e.g., JSON, image metadata).
Build custom web scraping solutions to collect real-time data from diverse sources, including financial reports, news, and publicly available datasets.
You’re not afraid to hop in and build a data integration for a customer, or take on other miscellaneous tasks.
Our goal is to provide the most accurate and easy to use forecasts to our customers, by leveraging all possible information on their industry. Foundation models for time series are changing this entirely. We can now pre-train a model on diverse temporal data across industries. Our users can rapidly interact with our models by changing inputs, providing context in natural language, and get immediate feedback on accuracy. Our users do not need to be data scientists or have an ML PhD to build and ship an accurate forecasting system for their use-cases.
Example use cases include demand forecasting for large furniture chains, for a restaurant group, and revenue forecasting in the gaming industry.
The founders are both ML PhDs who have built forecasting and ML systems at JP Morgan, Amazon, Google, Bloomberg, and Sonos in the US.
We are a global company that happens to be HQed in Paris. Get the best of both worlds — Silicon Valley work ethic and ambition in the center of Paris.