Eka Robotics Raises $150 Million for AI Robot Platform

The Cambridge-based company plans to scale its vision-force-action model for complex robotic manipulation tasks.

Updated on Oct. 24, 2026 in Robotics

Eka Robotics Raises $150 Million for AI Robot Platform

Live Poll

Do you believe robots should master physics instead of relying on human data for training?

Cambridge-based Eka Robotics has secured $150 million in Series A funding to scale its robotic manipulation technology. Founded in 2024, the company is currently developing a foundation model for robotic tasks using reinforcement learning.

Why it matters

This capital infusion allows the company to transition its research into broader customer deployments. The firm is competing to establish foundational AI models that can generalize physical tasks across diverse environments.

Robots utilizing the company's AI architecture perform approximately 1,500 operations per hour. This performance is achieved through simulation learning that trains a vision-force-action model based on reinforcement learning—a method where an algorithm learns to achieve a goal by interacting with an environment to maximize a reward.

The players

Eka Robotics

A Cambridge-based startup focused on developing foundation models for physical robotic manipulation using reinforcement learning.

Andreessen Horowitz

A prominent venture capital firm that invests in early-stage and growth-stage technology companies, specializing in AI and software infrastructure.

The details

Eka Robotics operates by synthesizing robotic experience through simulated environments, which accelerates training compared to real-world data collection. The system integrates visual perception with tactile force feedback to guide physical actions. This vision-force-action model is designed to handle complex manipulation tasks that typically require high-frequency control loops.

2024: Eka Robotics was founded in Cambridge, Massachusetts.
Week of October 24, 2026: The company announced the completion of its $150 million Series A funding round.

The Tech Race

Eka Robotics joins an intensifying competition to build universal foundation models capable of governing physical robotic actions. This development reflects the industry-wide push to move beyond hard-coded scripts toward AI that can learn and adapt to varied physical environments.

The firm plans to use this capital to increase its commercial deployments, potentially affecting logistics and manufacturing workflows in the near term. Potential users should monitor future customer case studies to see how the system handles real-world variability versus simulated training.

The takeaway

The company is betting that simulation-based training can solve the long-standing challenge of generalization in physical robotics. Observers should track upcoming pilot project benchmarks to verify if the 1,500 operations-per-hour speed translates from controlled settings to unpredictable, real-world customer sites.