ML Research Scientist at Guac (S23) $180K - $222K • 0.30% - 0.60% AI-powered demand forecasting for grocery New York, NY, US Full-time Will sponsor 1+ years About Guac Guac uses machine learning to do demand forecasting for grocers. We predict exactly how much of each product grocers will sell, on each day, at each of their stores. Our mission is to put an end to the millions of tons of food that goes to waste every single day due to bad inventory planning. About the role Skills: Torch/PyTorch, Python, Machine Learning ML Research Scientist - Guac At Guac, we're solving grocery food waste with predictive ML. We predict exactly how much of each product will sell, helping grocery retailers order/produce the perfect amount of inventory. The grocery industry is enormous (in the US, it accounts for 4% of GDP), and grocery food waste is one of the biggest drivers of cost for retailers, but also a huge problem for the climate and food security. Today, we're working with major supermarket chains in the US (you've probably shopped at some of them before), and we've scaled to 7-figures in ARR. We're also backed by leading investors such as Y Combinator, 1984 Ventures, Collaborative Fund, and angels from Instacart and Citadel Securities. We've brought together an exceptional team from Palantir, BCG, Oxford, Cambridge, and MIT to solve intellectually challenging problems and tackle food insecurity & waste with technology. We're looking for a talented ML Research Scientist in NYC to join our mission. About the Role As an ML Research Scientist at Guac, you'll be at the forefront of developing cutting-edge machine learning models that power our demand forecasting and inventory optimization systems. You'll tackle complex, multi-dimensional problems involving perishable goods, supply chain dynamics, and retail operations at unprecedented scale. Your work will directly impact millions of dollars in inventory decisions across major grocery chains, helping reduce food waste while ensuring customers have access to fresh products. This is a highly experimental role where intellectual curiosity and methodological rigor will drive measurable business outcomes. Core Responsibilities: Advanced Forecasting Models: Design and implement state-of-the-art ML models for demand forecasting at store, chain, and warehouse levels, accounting for seasonality, promotions, weather, and complex interaction effects Operations Research: Develop optimization algorithms for multi-supplier sourcing decisions, case size optimization, recipe planning, and inventory allocation across distribution networks Predictive Analytics: Build sophisticated models for current stock prediction, shelf-life optimization, and spoilage prevention using real-time data streams Research & Experimentation: Lead independent research initiatives to push the boundaries of what's possible in retail forecasting, with freedom to explore novel approaches and methodologies Model Validation: Design and execute rigorous A/B testing frameworks to validate model improvements and measure real-world impact on waste reduction and customer satisfaction Technical Environment: You'll work with our modern ML infrastructure built on Dagster for data orchestration, Dask and Coiled for distributed computing, and PyTorch and XGBoost for model development. Our data spans millions of SKUs, hundreds of stores, and billions of transactions. Success Metrics: Your primary measure of success will be forecast accuracy improvements that translate directly to reduced food waste and increased profitability for our retail partners. We expect you to consistently identify and implement improvements that move key business metrics. About You Required: 4+ years of experience in applied machine learning research, preferably in forecasting, optimization, or operations research Deep expertise in time series forecasting, with hands-on experience in PyTorch, XGBoost, or similar frameworks Strong mathematical foundation in optimization, probability, and statistics Proven track record of taking research ideas from conception to production deployment Highly Desirable: Experience with inventory optimization, supply chain analytics, or retail forecasting Background in operations research, particularly multi-stage optimization and stochastic programming PhD in a quantitative field (Machine Learning, Operations Research, Statistics, Economics, or related) or equivalent industry experience Experience with distributed computing frameworks (Dask, PySpark, etc.) Publication record in top-tier ML/OR conferences or journals Experience working in high-growth startup environments where research directly impacts business outcomes What We Offer Impactful Research: Work on problems that directly address climate change and food security Intellectual Freedom: Significant autonomy to explore novel approaches and drive research directions Compensation: $180k-$220k base + competitive equity Fully employer-paid healthcare (medical, dental, and vision) Unlimited vacation days Fully covered food expenses in the office (lunch/dinner) Free Equinox membership Conference & Research Budget: Support for attending top ML/OR conferences and continuing education Our Tech Stack ML & Research: PyTorch, XGBoost, Scikit-learn, NumPy, SciPy Data & Orchestration: Dagster, Dask, Coiled, Pandas, BigQuery Backend: Python, FastAPI Cloud & Infrastructure: GCP, Postgres, Terraform, Docker Frontend: React, Next.js, TypeScript Technology Ops: GCP (GCS, Cloud Run, Compute Engine mainly), FastAPI, Dask ML: PyTorch, XGB, OpenAI embeddings, SHAP, + more Codebase: Python