Data Scientist; 3yrs+ (part-time, immediate)

Job Description

We are hiring a Data Scientist (part-time) with a few years of experience. Expectation: about 15-20 hours per week for 3mths. Can turn into full-time for a strong performer. Join us and help us build the models that will define how resources are planned across Warehouses all over the US. You’ll lay the ML foundation to scale across thousands of warehouses and millions of packages. About Us Inveno is a US-based startup revolutionizing warehouse operations with intelligent resource planning powered by specialized AI. We’re on a mission to eliminate manual inefficiencies in warehouse management by building custom models that deliver smart, data-driven planning. What Problem Are We Solving — and Why? Today, 75% of US warehouses rely on manual shift planning — determining where staff go, which trucks unload where, which items get processed when — all without automation. This results in over $1 million in annual inefficiencies per warehouse. We are here to change that. By harnessing operational data and developing custom AI and optimization models, we help warehouse managers plan with precision, speed, and scale. Who’s Building This? We’re a founding team of 3: A warehouse operations expert with 8 years of experience at top logistics startups in the US and India. A CEO who’s built and exited 3 companies and led teams at Alibaba Group. A CTO who scaled engineering and systems at Bangladesh’s top-funded startup. Our Progress So Far Our V1 product is currently in development. We’ve already secured a paid contract with the global brand house with other major brands in the pipeline (including top US retailers) Key Responsibilities Lead Optimizer Development: Architect and implement our core optimization engine using tools like Google’s Math-OPT, NVIDIA cuOpt, or OR-Tools. Focus on improving labor planning, dock assignments, and wave planning to ensure on-time fulfillment. Build Machine Learning Models: Create and deploy supervised learning models to support and enhance the optimizer using complex real-world warehouse data. Shape LLM Interaction Layer: Collaborate with engineering to design workflows where LLMs convert user input into operational rules that feed into the optimizer. Establish MLOps Best Practices: Set up model versioning, deployment, monitoring, and experimentation pipelines that ensure robust and scalable ML infrastructure. Qualifications & Skills (Must-haves) Experience designing and deploying constraint-based optimization models using tools like Google OR-Tools, Gurobi, CPLEX, or similar. Advanced Python skills with fluency in the data science stack: NumPy, pandas, scikit-learn, etc. Strong foundation in supervised ML, including feature engineering, model selection, and validation techniques. Proven ability to translate complex operational problems into solvable mathematical or ML models. Bachelor’s or Master’s degree in Computer Science, Operations Research, Statistics, or a related quantitative field.