This simulation startup wants to be the Cursor for physical AI

2026-04-16 · Source: Robotics News | TechCrunch · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Robotics & Autonomous Systems, Software Development & Engineering · Depth: Intermediate, medium

Summary

Antioch, a New York-based startup, has raised an $8.5 million seed round, valuing the company at $60 million, to develop simulation tools for robot developers. The funding round was led by A* and Category Ventures, with participation from MaC Venture Capital, Abstract, Box Group, and Icehouse Ventures. Founded in May of last year by Harry Mellsop and four co-founders, including Alex Langshur, Michael Calvey, Collin Schlager, and Colton Swingle, Antioch aims to address the "sim-to-real gap" in robotics. This gap refers to the challenge of creating virtual environments realistic enough for robots trained within them to operate reliably in the physical world. The company's product allows robot builders to create multiple digital instances of their hardware, connecting them to simulated sensors to mimic real-world data, enabling testing of edge cases, reinforcement learning, and training data generation. Antioch's current focus is on sensor and perception systems for automated cars, trucks, farm machinery, and drones, working with models from Nvidia and World Labs.

Key takeaway

For robotics engineers and AI scientists developing autonomous systems, Antioch's simulation platform offers a critical solution to the "sim-to-real gap." You should explore integrating high-fidelity simulation tools to reduce reliance on costly physical testing and accelerate development cycles. This approach allows for efficient testing of edge cases and generation of training data, potentially enabling faster deployment of robust physical AI applications.

Key insights

High-fidelity simulation is crucial for scaling physical AI and bridging the gap between virtual training and real-world robot deployment.

Principles

• Physical AI needs scalable data.
• Simulation reduces real-world data costs.
• High-fidelity physics is critical.

Method

Antioch's platform spins up digital hardware instances connected to simulated sensors, mimicking real-world data for testing edge cases, reinforcement learning, and generating training data.

In practice

• Test robot designs in simulation.
• Generate training data virtually.
• Evaluate LLMs in simulated environments.

Topics

• Physical AI
• Robotics Simulation
• Sim-to-Real Gap
• Autonomous Systems
• Sensor Perception

Best for: Machine Learning Engineer, Computer Vision Engineer, AI Scientist, Robotics Engineer, AI Engineer, Research Scientist

Related on AIssential

• Training a Unitree G1 to Walk w/ Reinforcement Learning — Achieving sim-to-real transfer for robot locomotion requires meticulous sim-to-sim fidelity and addressing hardware-specific observation gaps.
• Building brains for bulldozers — Simulation is essential for scaling robotics, complementing real-world data for robust development.
• ABB, Nvidia Partner to Deliver Physical AI at Scale — ABB and Nvidia's partnership aims to bridge the sim-to-real gap in robotics through hyper-realistic simulation and AI model training.
• What is Physical AI? How Robots Learn & Adapt in Real Life — Physical AI integrates perception, reasoning, and action in the physical world, driven by advanced models and efficient compute.
• [D] Sim-to-real in robotics — what are the actual unsolved problems? — Sim-to-real transfer faces practical hurdles in robotics, despite advanced simulation platforms.
• Intelligent Robots in 2026: Are We There Yet? with Nikita Rudin - #760 — Bridging the sim-to-real gap and integrating perception are key challenges for deploying autonomous robots.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Robotics News | TechCrunch.