AI-powered dual-arm robotics for intelligent manufacturing: R&D and real-world deployment - BC-1009
Project type: ResearchDesired discipline(s): Engineering - computer / electrical, Engineering, Engineering - mechanical, Computer science, Mathematical Sciences
Company: TP7 AI Robotics
Project Length: 6 months to 1 year
Preferred start date: As soon as possible.
Language requirement: English
Location(s): Vancouver, BC, Canada
No. of positions: 1 - 2
Desired education level: Undergraduate/BachelorMaster'sPhD
Open to applicants registered at an institution outside of Canada: No
About the company:
TP7 AI Robotics is a Physical AI company building proprietary Neuro-Symbolic AI for industrial dual-arm robots deployed in high-mix manufacturing environments. Our flagship system, Ada, enables robots to understand and execute complex, variable assembly tasks without task-specific training data — achieving SKU onboarding in under 5 minutes and language-to-motion planning in under 3 minutes. We deploy on a Robotics-as-a-Service model, making advanced automation accessible to manufacturers who cannot support large in-house robotics teams. TP7 is backed by Harvard Innovation Labs, NVIDIA Inception, and MassRobotics, with support from NRC Canada and the Canadian Robotics Council. Our mission: build robots that work reliably in the real world, not just in controlled lab conditions.
Describe the project.:
The intern will work directly with TP7’s CTO on R&D for Ada, TP7’s Neuro-Symbolic Physical AI system deployed on industrial dual-arm robots in high-mix manufacturing environments. Ada combines symbolic reasoning with learned perception to enable robots to interpret natural language instructions, plan multi-step manipulation sequences, and execute precise physical tasks — without task-specific training data. Key R&D activities will include: (1) designing and improving perception pipelines for 3D object detection, pose estimation, and scene understanding in unstructured factory settings; (2) developing and evaluating motion planning and trajectory optimization algorithms for dual-arm manipulation; (3) hardware integration work including interfacing with robot controllers, sensors, and communication protocols such as Bluetooth and other embedded connectivity layers; (4) testing and validating system performance in real and simulated deployment environments; and (5) contributing to data pipelines, evaluation frameworks, and internal tooling. Expected deliverables include working code contributions to TP7’s core robotics stack, technical documentation of implemented systems, and participation in real-world pilot deployments with manufacturing customers. The intern will gain hands-on experience bridging AI research and physical deployment — working with actual robotic hardware, not simulations alone.
Required expertise/skills:
Strong programming skills in Python are required; experience with C++ is a strong plus, particularly for real-time robotics and hardware interfacing. Candidates should have familiarity with robotics frameworks (ROS/ROS2) and at least one of: computer vision libraries (OpenCV, Open3D), deep learning frameworks (PyTorch, JAX), or motion planning libraries (MoveIt, OMPL). Experience with hardware communication protocols — including serial (UART), CAN bus, Bluetooth (BLE), or similar embedded interfaces — is a meaningful advantage for hardware integration work. Relevant mathematics includes linear algebra (matrix operations, SVD, spatial transformations), probability and statistics (Bayesian inference, Gaussian distributions), and optimization (gradient-based methods, convex optimization). Physics knowledge should include rigid body dynamics, kinematics and inverse kinematics, and an understanding of forces and torques in manipulation contexts. Familiarity with 3D geometry, point clouds, or SLAM is a strong advantage. Prior experience with physical robotic systems, embedded hardware, sim-to-real transfer, or manipulation research is highly valued but not required.
