Case Study: Guac reduces food waste with scalable demand forecasting using Coiled

How Guac scales demand forecasting to reduce food waste

Guac, an AI startup helping food retailers forecast demand to reduce food waste, faced a significant technical challenge when onboarding enterprise-scale customers. Their existing infrastructure, built on pandas, was unable to efficiently process the exponentially larger datasets from these new clients, which included rewrite, leading them to the vendor Coiled.

By implementing Coiled's platform, Guac easily scaled their Dask-based workflows to run distributed computations across multiple cloud machines with minimal code changes. This solution allowed them to effortlessly handle massive ETL jobs and computationally intensive machine learning forecasting for retailers of any size. As a result, Coiled eliminated scale as a barrier to growth, compressing customer onboarding to just one or two months and allowing Guac's engineering team to focus on improving their forecasting models rather than maintaining infrastructure.

Guac

Jack Solomon

CTO and Co-founder