.png){kind=logo}
Anay Dongre
Junior Machine Learning Engineer specializing in GenAI and LLM fine-tuning
Connect with Anay
Anay already has a relationship with Reval, so a warm intro from us gets a much better response than cold outreach.
Recommended
Already have an account?
About
Robotics software engineer focused on hard real-time autonomy for legged robots, building a quadruped navigation stack that combines vision SLAM with MPC and maintains a deterministic 500Hz control loop. Deep performance optimization experience across CUDA (sub-2ms perception latency), ROS 2/DDS real-time tuning, and motion planning (cut 500ms spikes to sub-5ms). Also designed distributed ROS 2 + Zenoh communications between quadrupeds and aerial drones and validated robustness under lossy wireless conditions.
Experience
Education
Key Strengths
- Built real-time quadruped navigation stack integrating vision SLAM + MPC for 500Hz deterministic control
- GPU/perception optimization: reduced perception-to-estimation latency from 12ms to deterministic sub-2ms (zero-copy + fused CUDA kernels)
- Real-time ROS 2 architecture for deterministic state management and high availability (lifecycle nodes, multi-threaded executor, DDS tuning)
- Motion planning performance debugging: eliminated 500ms spikes by refactoring A* to Jump Point Search with pre-allocated memory pool (sub-5ms planning, 200Hz local planner)
- Distributed multi-robot comms: ROS 2 + Zenoh bridge with Protobuf serialization achieving sub-10ms inter-robot latency
- Robustness validation under adverse networks: simulated 20% packet loss/jitter with Linux tc; failover to local autonomy within 50ms
- DevOps for robotics: Docker + GitHub Actions CI with headless Gazebo regression tests; resolved CUDA version drift and caught sensor-fusion race condition
Discover more candidates like Anay
Search across thousands of pre-screened, high-quality, high-intent candidates on Reval.
Search Talent