- Company Name
- Chewy
- Job Title
- AI Research Scientist Intern
- Job Description
-
**Job Title**
AI Research Scientist Intern – Physical AI
**Role Summary**
PhD research intern working on fleet intelligence for heterogeneous humanoid and mobile robots. The role focuses on developing algorithms for shared spatial understanding, large‑scale decision-making, and multi‑robot coordination, with an emphasis on simulation‑first experimentation and progressing findings toward lab validation and publication.
**Expectations**
- Execute end‑to‑end research (problem definition, baseline development, implementation, evaluation) under guidance of senior researchers.
- Produce rigorous, reproducible results that meet publishable quality.
- Integrate research outputs into a broader fleet‑orchestration workflow and collaborate with AI, robotics, and systems engineering teammates.
- Communicate findings through manuscripts, technical reports, and internal presentations.
**Key Responsibilities**
- Formulate research problems and design baseline approaches in one primary area: collaborative SLAM/estimation, task allocation/scheduling, motion planning MAPF/routing, fleet safety supervision, or learning‑assisted decision‑making.
- Implement algorithms in Python/C/C++/ROS/ROS2/Julia and deploy them in robotics simulators (MuJoCo, Isaac Sim, PyBullet, Gazebo).
- Build simulation‑first experiment suites, define evaluation metrics (throughput, latency, robustness, safety‑event rate, localization error, etc.), run ablation studies, and analyze outcomes.
- Develop datasets, benchmarks, and reusable evaluation harnesses, maintaining strong reproducibility standards.
- Translate results into laboratory validation prototypes when feasible.
- Author manuscripts for peer‑review conferences/journals and prepare technical documentation for internal stakeholders.
**Required Skills**
- PhD‑level expertise in robotics, AI, or related field with focus on perception/estimation, control, optimization, multi‑agent systems, or machine learning.
- Deep knowledge in at least one of the following: multi‑agent estimation/mapping (C‑SLAM, semantic mapping, sensor fusion), multi‑agent task allocation/scheduling (LSTA), MAPF/routing, control + state estimation (MPC, filtering), ML for decision‑making (RL/IL/model‑based).
- Strong programming in Python, C/C++, or Julia; experience with ROS/ROS2.
- Proficiency in robotics simulation environments and ability to prototype and debug algorithms.
- Experimental and analytical reasoning skills, capable of designing, executing, and iterating on evaluation studies.
- Demonstrated publication record in top venues (NeurIPS, ICLR, ICML, ICRA, IROS, CDC, etc.) or clear evidence of research capability.
- Excellent written communication and presentation skills.
**Required Education & Certifications**
- Currently enrolled in a PhD program in Robotics, Electrical Engineering, Mechanical Engineering, Computer Science, Applied Mathematics, or a closely related discipline.
- Coursework or research emphasis on perception, estimation, control, optimization, multi‑agent systems, robotics, or artificial intelligence.