Greg Shklovski
Deploying Humanoid Robots @ Foundation · Robot Learning Research · Startups
I build practical robot learning systems and fast demos for startups. My work spans on-device perception, imitation learning for manipulation, and human-in-the-loop interfaces that make robots useful in the real world.
Projects
Selected projects and demos
Envision
2025Gesture interface device + SDK with on-device Coral TPU acceleration; event-driven APIs for fine-grained hand inputs.
Hungry Monkey
2025YC x Supabase AI hackathon – “Yelp for Holiday Hours.” FastAPI backend + React frontend. See Loom demo in outline.
Gesture Canvas
2024A dynamic canvas for sketching ideas using hand-tracking gestures to draw and manipulate elements in real time.
HandCode
2024Prototype that lets you code with wrist gestures, mapping movement patterns to editor actions.
git.ar
2023Spatial computing tool that visualizes git commit graphs in AR, supports hand-tracked merging, and uses LLMs for diff summaries.
YapTrack
2023Real-time meeting guide that keeps discussions on track and turns ideas into follow-ups.
Basket
2022Collaborative grocery delivery experiment.
Clock In
2022Body-mounted dash cam to protect workers in hazardous conditions.
Skills
Experience
- Deploy humanoid robots to customers; integrate hardware, perception, and control into reliable demos and pilots.
- Close the loop between research and deployment with a focus on safety, reliability, and iteration speed across perception, control, and demos.
- Trained and deployed adaptive intention recognition enabling a UR-3e arm to assist with ambiguous manufacturing tasks.
- Built data collection, labeling, and evaluation pipelines; delivered baselines for intent and state-change recognition.
- Built an ETL pipeline powering the Factory Platform (Python, GraphQL).
- Improved data reliability and schema evolution for manufacturing workflows.
- Shipped two full-stack apps (C#/.NET + Angular).
- Implemented auth, database, file system, and calibration workflows.
Publications
Awards
Media
Get in touch
I’m excited to collaborate with teams pushing embodied AI into the real world. Reach out on LinkedIn or explore more demos on Devpost.