DataraAI builds robots to cable AI servers, joins Arm Total Design

DataraAI is a California-based company focused on simulation and training data for manufacturing automation, especially AI server and rack assembly and tasks involving deformable components and physical contact. Its current effort targets the cabling stage, where cables change shape as they are manipulated and successful connections can be difficult to verify visually. DataraAI says cabling can become a bottleneck because thousands of connections may still be routed and terminated manually even when other parts of the process are already automated.

Rather than generating a simulated environment first and then transferring a robot's learned behavior to the physical line, DataraAI calibrates its simulations using data collected from real operations, including RGB-D vision, force and torque measurements, tactile sensing and motion data. This is intended to address the so-called Sim2Real gap, where robotic systems that perform well in simulation run into trouble with the physical variability of real components and environments. For cable insertion, vision can locate the cable and connector, while force and tactile information can help establish whether the connector has actually been seated correctly. CEO and founder Durgesh Srivastava argues that self-driving technology reached autonomy by driving real miles and building the simulator from them, and that industrial manufacturing needs the same order of operations. He adds that rack output is gated at the cabling station and that manufacturers are asked to commit capital to automate a task no vendor has proven on their hardware.

The company has joined Arm Total Design for Physical AI, an ecosystem spanning AI models, software, sensors, silicon and development tools, and is working with Arm on its Robotics Capability Framework, an initiative to establish a common way of describing and comparing robotic system capabilities. Arm's Dermot O’Driscoll says scaling physical AI from innovation to production requires systems that operate reliably in real-world complexity and variability, and that DataraAI brings simulation accuracy expertise toward that goal. The stakes appear to hinge on whether DataraAI's calibration-first approach holds on a manufacturer's own rack and connectors before any purchase, since the value case rests on proving the task on real hardware. For production and procurement teams, that proof point, rather than simulation results alone, is likely to determine whether cabling automation moves from pilot to capital commitment.