Robotic grocery delivery
As pandemic forced us to do more groceries online delivery apps have become a popular and convenient way to go shopping. With projected growth to 22% by 2025 online delivery apps break all records, nevertheless, they are still not making money. Inefficient use of human labor leads to high costs and does not allow them to reach sustainable profitability even in the long term. The only way to profitability is a highly integrated solution unifying concepts of dark stores and autonomous robotic delivery.
We are building a scalable robotic solution with a focus on suburbs and districts with family homes. Our main mission is to offer 30-minutes free grocery delivery for everyone.
Robotic grocery delivery for suburbs
End-to-end Deep Learning Approach
Most of the training is done in computer simulations using reinforcement and imitation learning
System that learns by itself to understand useful road features
Autonomous droids need better brains, not more hardware. Unlike traditional, unscalable approaches that use complex maps, scripted rules and, expensive sensors, we are building an end-to-end system that drives anywhere by observing and understanding its surroundings. Just like a human.
Robust to the new scenes and environments
Rule-based systems that try to predict every edge case might produce encouraging results in pre-fitted narrow scenes. However, in the long term, only an end-to-end approach can lead to maximum efficiency.
No need for expensive hardware (lidar, radar, etc.) and HD maps
Scalable neural models - this is what determines the agility of autonomous vehicles, not the number of lidars and radars. Our system shows robust results with couple low-res cameras.
Easy to scale, easy to feed more data
There is no more need to train the model exclusively in a real-world environment. More than 90% of the training can be done in a computer simulation using reinforcement and imitation learning. Team
We are passionate about the future of mobility
and machine learning.
Mike Melanin / CEO
  • Young Entrepreneur of the Year 2017 by E&Y;
  • The Founder of (Business Intelligence solution used by over 30,000 companies and Startup of the Year in Russia);
  • Backed by Slack, 500startups, Eniac VC, BetaWorks and Eric Schmidt's Innovation Endeavors;
  • Graduated from Bauman University / Robotics and Complex Automation.
Kostya Selivanov / CTO
  • 10+ years experience in machine learning, 6+ years in deep and reinforcement learning;
  • Lead machine learning engineer at Orbitalinsight (backed by Sequoia, GV for more than $100M);
  • Lead machine learning researcher at Salariat (acquired by Genesys).
Feel free to write, call and visit us.
We really love communicating with our clients.


© Nopilot Technologies.