Job Description

Terafac is seeking a

Machine Learning Engineer

to join our team focused on making a transformational AI-Vision platform for Industrial Robots. The ideal candidate will have exposure to building data pipelines, backend services, and training ML models. The position will involve taking these skills and applying them to solve for some of the most crucial & most exciting technology challenges of our generation.

About Terafac

Terafac is where AI meets the real world.

We're building an AI-Vision platform that enables industrial robots to see, think, and adapt like skilled human workers - transforming how physical work gets done across global manufacturing.

Operating at the intersection of AI, computer vision, robotics, and manufacturing, Terafac is addressing global challenges such as labour shortages and the need for scalable automation. Our systems are already live in production environments, with active scale-up across India and international markets.

At Terafac, you'll work on some of the hardest, most meaningful problems in AI, Computer-Vision and robotics, with deep hands-on exposure and ownership in a fast-paced, high-impact startup shaping the future of manufacturing.

Key Responsibilities

Design, train, evaluate, and deploy perception models with a strong emphasis on segmentation, localization, and spatial understanding in real-world environments Own the end-to-end ML lifecycle: data collection and labeling strategy, model development, evaluation, deployment, monitoring, and iteration Work closely with data and ops teams on large-scale data collection and augmentation pipelines, including handling long-tail and failure cases Develop and validate data augmentation strategies (geometric, photometric, synthetic, domain randomization) to improve robustness and generalization Analyze model failures (e.g., over/under-segmentation, drift, confidence collapse) and design diagnostics, safeguards, and fallback strategies Experiment with model compression and distillation (teacher-student setups) for efficient deployment on constrained hardware Collaborate with robotics, backend, and systems teams to ensure ML models integrate cleanly into production pipelines Make principled trade-offs between accuracy, latency, compute, reliability, and maintainability, and clearly communicate assumptions and limitations

Minimum Qualifications

Bachelor's or Master's degree in Computer Science or related fields, and/or equivalent hands-on machine learning experience 2-5 years of practical experience building, training, and evaluating machine learning models in real projects Hands-on experience with computer vision models, especially segmentation (semantic and/or instance) Strong fundamentals in probability, statistics, linear algebra, and optimization Solid experience with modern ML frameworks (e.g., PyTorch) and strong Python programming skills Ability to reason about data distributions, noise, bias, and failure modes in ML systems

Preferred Qualifications

Experience working on a production perception stack, not just isolated models Exposure to Vision Transformers (ViTs) or hybrid CNN-Transformer architectures Hands-on experience with knowledge distillation or teacher-student training approaches Experience collaborating with teams on data collection, labeling, and augmentation at scale Familiarity with vision-language models or Vision-Language-Action (VLA) systems, or strong interest in multi-modal learning Experience deploying ML models in production environments (cloud, edge, or on-device) and working under real-world constraints

Job Details

Job Type:

Full-time

Location:

Chandigarh, Sector 25 (Work from Office)

Contact Details

Send your resume to: hiring@terafac.com

cc: simar-sethi@terafac.com

Contact Number:

+91-8264445150

Job Types: Full-time, Permanent

Pay: From ?1,000,000.00 per year

Benefits:

Health insurance Paid sick time Paid time off Provident Fund
Ability to commute/relocate:

Chandigarh, Chandigarh: Reliably commute or planning to relocate before starting work (Required)
Experience:

Machine learning: 2 years (Required)
Work Location: In person