Harsh Pant
Operations Research Engineer @knuspr.de
Signup · Get unlimited contacts
WORK HISTORY
Operations Research Engineer @knuspr.de
Munich, DE
Engineered constraint programming model using OR-Tools CP-SAT solver to optimize bakery production scheduling across multiple ovens, achieving 5% shrink reduction and 20% decrease in oven usage while maintaining product freshness standards• Developed geospatially-aware quantile regression model for package delivery handling time prediction incorporating UberH3 hexagonal grid features, reducing MAE by 25% and enabling robust operational planning under demand uncertainty• Built production-grade time-series forecasting system for truck loading duration, achieving 3-minute per-vehicle reduction that enabled €1M annual savings through optimized route planning and reduced fleet requirements• Architected end-to-end ML pipelines using ZenML orchestration framework and Docker containerization for automated feature engineering, model training, and deployment to production with monitoring and versioning
EDUCATION
The LNM Institute of Information Technology
Bachelor of Technology - BTech, Computer Science
ABOUT HARSH PANT
I turn messy supply chain problems into elegant solutions using AI and math. And yes, it\'s as fun as it sounds!Here\'s what I do: I\'ve spent the last 7+ years building AI and machine learning systems that actually work in the real world — not just in notebooks. Think warehouse robots knowing exactly where to go, trucks loading faster, and bakeries baking just the right amount of bread. The results? Companies save millions, operations run smoother, and I get to solve really interesting puzzles every day.What I Do:My sweet spot? Taking complex math problems and turning them into production systems that people actually use. I work on things like predicting when your packages will arrive, figuring out the smartest way to arrange products in warehouses, and optimizing delivery routes for thousands of drivers. It\'s a mix of hardcore algorithms, machine learning, and making sure everything runs smoothly when deployed.Core Expertise:Mathematical Optimization — I live and breathe Mixed Integer Programming, Constraint Programming, and all sorts of metaheuristics (genetic algorithms are my favorite). Tools of choice: Gurobi and OR-Tools.AI & Machine Learning — From time-series forecasting to geospatial analytics, I build models that predict what\'s coming next. Big on MLOps too (ZenML and Docker are my daily drivers).Real-World Problems — Supply chain optimization, last-mile delivery, production planning, inventory management, and figuring out where to put warehouses. Basically, making logistics less chaotic.Recent Achievements:Helped a bakery waste 5% less food and use 20% less oven time (because nobody likes burnt bread or energy waste)Built a geospatial AI model that predicts delivery times 25% more accurately (your pizza arrives when it says it will!)Deployed an allocation system across warehouses that cut sorting time in half (that\'s a LOT of saved hours)Optimized a delivery network that saves over €1M every year in transportation costs (math pays off!)Technical Stack:Python | Gurobi | OR-Tools | Docker | ZenML | AWS | SQL | NetworkX | UberH3 | GitCurrently:Right now, I\'m based in Munich working at Knuspr, where I get to solve real-time optimization problems in the e-grocery world. Think of it as making sure fresh groceries get to your door as efficiently as possible.Want to chat about optimization, AI, or how to make supply chains less painful? I\'m always up for a good conversation about technical challenges, interesting problems, or just geeking out about algorithms. Feel free to reach out!
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.