Computer Vision Engineer

 Location: Bangalore (ARTgarage)

About us 

idle Robotics is a Bengaluru-based startup with the ambitious mission to become the intelligence layer powering all autonomous systems. We believe the future of AI is physical, and the first step is giving machines the ability to perceive the world. We are building a foundational layer of visual intelligence that is, in effect, the "visual cortex" for robots, allowing them to perceive, navigate, and act intelligently. Our dual-use approach means you will contribute to high-impact work, from GPS-denied navigation for defense drones to scalable software for global industrial automation. If you are passionate about robotics computer vision and pushing the boundaries of Physical AI, we invite you to build the core technology with us.

Role Summary

As our Computer Vision Engineer , you will build classical vision pipelines, deep learning architectures, and foundation model adaptations for detection, segmentation, tracking, and 3D perception. You will optimize models for embedded and edge platforms, work with large datasets, and collaborate across robotics and system architecture teams to bring robust perception systems from prototype to field deployment.

Responsibilities

• Develop computer vision pipelines using image processing, feature extraction, and classical CV methods

• Build deep learning models for detection, segmentation, tracking, and 3D perception using CNNs, Transformers, and architectures such as YOLO, Faster RCNN, Mask RCNN, UNet, and DeepLab

• Fine-tune and adapt foundation models such as SAM, CLIP, and DINO

• Optimize model performance for embedded and edge compute platforms

• Design and manage datasets including cleaning, augmentation, and annotation using CVAT or Label Studio

• Run evaluation, profiling, and failure analysis on deployed models

• Collaborate with robotics and embedded teams to integrate perception outputs into navigation, control, and planning

• Maintain documentation, experiment logs, and deployment specifications

Minimum Qualifications

• Strong foundation in classical computer vision, geometry, and camera calibration

• Proficiency in PyTorch or TensorFlow

• Experience with YOLO, RCNN family models, UNet, or DeepLab

• Proficiency in Python and core libraries such as OpenCV, NumPy, SciPy, TorchVision, and Albumentations

• Strong math foundations including linear algebra, calculus, and probability

• Hands-on experience with dataset creation and annotation

• Ability to write clean and modular code and work collaboratively

Bonus Qualifications

• Experience with 3D vision, stereo, SLAM, or reconstruction

• Familiarity with self-supervised or vision-language models

• Experience using TensorRT, ONNX Runtime, or similar tools

• C++ proficiency for performance-critical CV modules

• Robotics experience integrating perception pipelines

Why Build With Us?

• Collaborate directly with IIT/IISc founders in a high-density engineering environment

• Develop dual-use technology for national defense (GPS-denied navigation) and industrial automation

• Solve complex "zero-to-one" problems in Physical AI and autonomous systems

• Receive substantial ESOPs and influence as a core team member

• Access competitive compensation, paid time off, and a growth-focused culture

ARTPARK @ IISc :  Innovation factory for next-gen robotics & AI
ARTPARK is India's leading deep-tech venture builder and incubator focused on robotics, connected autonomous systems, and AI. Leveraging our unique facilities and ecosystems, we strive to provide meaningful support to very early-stage startups building deep-tech products based in research. We are a nonprofit organization created by Indian Institute of Science (IISc, Bengaluru) with support from the Department of Science & Technology (Government of India) and the Government of Karnataka.