9+ years in product management. I build products at the intersection of spatial intelligence, autonomous systems and AI safety, ensuring agents perceive, reason, and act within strictly defined Operational Design Domains (ODD) in the real world.
My background in Physical AI goes back to 2016, when I worked with the iCub humanoid robot on visual recognition, semantic reasoning, and visual servoing systems. That early robotics experience shapes how I approach autonomous product development today.
SAR and Earth Observation AI pipelines for geospatial intelligence. Synthetic data generation using NVIDIA Omniverse Replicator and world foundation models for closing the sim to real gap. See pytorch-shield.
Deep RL foundations, multi-agent reinforcement learning, and coordinated autonomous operations using PettingZoo, Ray RLlib, and MAPPO. See multi-shield-europe.
Real-time safety monitoring for autonomous fleets with teleoperation trigger detection. Adversarial scenario generation, safety benchmarking, and applying UL 4600 and SOTIF (ISO 21448) standards to autonomous systems.
| Quarter | Project | Focus |
|---|---|---|
| Q1 | Operational Safety Monitor | Real-time safety monitoring dashboard for autonomous fleets — teleoperation trigger detection built on Waymax and Waymo Open Motion Dataset. |
| Q1 | SAR & EO AI Workflows | Synthetic Aperture Radar and Earth Observation AI pipelines — satellite data acquisition, processing, and analysis using UP42 Python SDK for geospatial intelligence workflows. |
| Q2 | Deep RL Foundations | Policy gradients, model based RL, robot learning |
| Q2 | Multi-Agent RL — Europe | Multi-agent reinforcement learning and coordinated autonomous operations using PettingZoo, Ray RLlib, and MAPPO. See multi-shield-europe. |
| Q2 | AV Safety Benchmark | Adversarial scenario generation and safety metrics |
| Q3 | Synthetic Data Generation | Domain randomized synthetic data pipelines for long tail edge case coverage using NVIDIA Omniverse Replicator and procedural scenario generation. |
| Q3 | World Model Benchmark | World foundation model evaluation and sim to real gap |
| Q4 | Deep RL for Robotics | Applying deep RL to robotic manipulation and locomotion — sim to real transfer, contact rich tasks, and embodied policy deployment. |
I run an AI augmented PM Operating System across three environments:
| Context | Tooling | Use Case |
|---|---|---|
| Personal / AI Builder | Claude Code (personalized) | GitHub projects, research synthesis, personal productivity |
| Enterprise PM | M365 Copilot | Product strategy, roadmaps, business models, stakeholder communication |
| Production Engineering | Amazon Kiro | Requirements driven implementation tracking and status visibility |
| Layer | Technologies |
|---|---|
| ML / RL | PyTorch, Stable Baselines3, PettingZoo, Ray RLlib |
| Robotics | ROS2, Gazebo, YARP, MuJoCo |
| Safety | UL 4600, SOTIF (ISO 21448) |
| Synthetic Data | NVIDIA Omniverse Replicator, Waymax, Procedural Generation |
| SAR / EO | UP42 Python SDK, Rasterio, GDAL, SentinelHub |