> Markdown version of [/jobs/ext/2934452-applied-scientist-one-mhs-software-controls-science](https://www.wearedevelopers.com/jobs/ext/2934452-applied-scientist-one-mhs-software-controls-science). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md). --- # Applied Scientist, One MHS - Software, Controls, Science - **Company:** Amazon.com, Inc. - **Location:** Boston, MA, United States - **Experience:** Experienced - **Salary:** $142,800.0 - $193,200.0 - **Contract:** Permanent contract - **Skills:** Artificial Intelligence, Artificial Neural Networks, Computer Vision, Business Software, Network Congestion, Python (Programming Language), Linear Programming, Machine Learning, Software Engineering, Reinforcement Learning, Pytorch, Deep Learning, Information Technology, AISTATS - **Published:** September 16, 2026 - **Apply:** https://www.jobmonkeyjobs.com/career/28023096/Applied-Scientist-One-Mhs-Software-Controls-Science-Massachusetts-Boston-7375 ## About the Role 2+ years of building machine learning models or developing algorithms for business application experience - PhD in Operations Research, Statistics, Applied Mathematics, Engineering, Computer Science or related field - Experience in optimization mathematics such as linear programming and nonlinear optimization - Knowledge of and proficiency in the use of Python scripting language - Experience Experienced with end-to-end ownership of major project deliverables - Experience with popular deep learning frameworks and RL tooling (e.g., PyTorch, d3rlpy, Ray/RLlib, Gymnasium, Stable-Baselines3, Isaac Gym/Omniverse) - Demonstrated experience developing and applying optimization or reinforcement learning solutions (e.g., MILP, constraint programming, stochastic programming, contextual bandits, deep RL) to real-world control, scheduling, or operation problems Preferred Qualifications - First-author publications at top-tier machine learning, operations research, or control venues (e.g., NeurIPS, ICML, ICLR, AAAI, AISTATS, CPAIOR, INFORMS Journal on Computing, or IEEE control and automation conferences) - Experience building a discrete-event simulator to train and evaluate operational policies, and calibrating it against historical data - Experience applying optimization or RL in a setting analogous to ours: production scheduling, real-time industrial control, robotics, material handling, industrial process or operations. - Experience deploying optimization or ML models to production at scale and partnering with engineering teams on inference, monitoring, and feedback loops ## Description As an Applied Scientist, you will collaborate closely with other scientists and engineers to bring optimization and sequential decision-making research to production. This role combines the scientific application of ML, and specifically optimization, RL, and sequential decision making, with software development engineering and a strong product focus. It will be your job to design, implement, and deploy novel decision policies and optimization models in both prototype and production environments, and to prove their impact through rigorous evaluation and simulation before scaling them across the fleet., * Own the research and development of optimization and sequential decision-making solutions spanning constraint programming, stochastic and robust optimization, contextual bandits, and reinforcement learning for real-time MHE control and scheduling optimization in a production environment. * Formulate fulfillment operations and manufacturing scheduling problems (production scheduling, resource allocation, sorter optimization, throughput and congestion control) as optimization or sequential decision-making problems, and design multi-objective functions that balance competing operational objectives such as on-time delivery, utilization, changeover cost, and schedule stability. * Build and leverage high-fidelity simulation and emulation environments for safe offline training, policy validation, and transfer to live systems before fleet-scale deployment. * Collaborate across multiple science and engineering teams to integrate policies into production planning and real-time control systems, including monitoring, guardrails, and staged rollout. * Communicate results and their limitations clearly in writing to technical and business audiences, and contribute to the team's external research presence through publication where the work merits it. About the team Amazon is building next generation software, hardware, and processes that will run our global network of fulfillment centers that move millions of units of inventory, and ensure customers get what they want when promised. The Science Software team in the One MHS organization unlocks Material Handling Equipment (MHE) innovation through a multiplicity of disciplines within Artificial Intelligence (AI) and applied science, including Optimization, Reinforcement Learning, classical Machine Learning, statistical modeling, Computer Vision (CV), and Physics-Informed Neural Networks (PINNs). The team is dedicated to building self-optimizing fulfillment centers, developing the models that drive real-time, building-wide orchestration of MHE. We conduct experiments, develop models, and apply machine learning (ML) at scale to optimize throughput, flow, merge, and congestion control, and to improve operational performance across the fulfillment network. ## Related Videos - [Getting Started with Machine Learning](https://www.wearedevelopers.com/videos/260-getting-started-with-machine-learning) - [Geometric deep learning for drug discovery](https://www.wearedevelopers.com/videos/264-geometric-deep-learning-for-drug-discovery) - [Navigating the AI Revolution in Software Development](https://www.wearedevelopers.com/videos/1266-navigating-the-ai-revolution-in-software-development) - [Photonic Computing: Programming a New Class of AI Accelerators (incl. Live Coding)](https://www.wearedevelopers.com/videos/100196-photonic-computing-programming-a-new-class-of-ai-accelerators-incl-live-coding) - [Effective Machine Learning - Managing Complexity with MLOps](https://www.wearedevelopers.com/videos/185-effective-machine-learning-managing-complexity-with-mlops) - [Serverless deployment of (large) NLP models ](https://www.wearedevelopers.com/videos/158-serverless-deployment-of-large-nlp-models) ## Related Articles - [How to Become an AI Engineer](https://www.wearedevelopers.com/magazine/331-how-to-become-an-ai-engineer) - [MLOps – What’s the deal behind it?](https://www.wearedevelopers.com/magazine/125-mlops-what-s-the-deal-behind-it) - [What Industries Outside of AI Are Hiring The Most AI Experts?](https://www.wearedevelopers.com/magazine/98-what-industries-outside-of-ai-are-hiring-the-most-ai-experts) - [MLOps And AI Driven Development](https://www.wearedevelopers.com/magazine/82-mlops-and-ai-driven-development) - [From Prototype to Production: Build AI Agents with This Free 4-Course Learning Path](https://www.wearedevelopers.com/magazine/655-from-prototype-to-production-build-ai-agents-with-this-free-4-course-learning-path) - [Navigating the AI Shift](https://www.wearedevelopers.com/magazine/629-navigating-the-ai-shift)