Computer Vision Engineer

Summary: Hands-on Computer Vision Engineer to own CV problems end-to-end for industrial AI applications. Responsibilities include model development, edge deployment, inference optimization, and production monitoring for complex vision challenges in heavy industry.

About Ripik.AI

Ripik.AI is an Accel-backed Applied AI company building computer vision and process-optimisation agents for the world's largest industrial enterprises across steel, cement, aluminium, chemicals, pharma, and power. Our Vision AI platform acts as an automated pair of eyes on the shop floor — monitoring materials, equipment, and processes 24/7 with 95%+ accuracy, eliminating human error, and delivering measurable gains in throughput, yield, energy efficiency, and safety.

We work with marquee customers including Tata Steel, JSW, ArcelorMittal, Vedanta, Godrej & Boyce, Grasim, Holcim, and Jindal Steel, and are scaling globally across India, the Middle East, Europe, and North America. We are one of the few Indian AI product start-ups to be a partner to GCP, Azure, and AWS, and the AI partner of choice for CII, ICC, and NASSCOM.

The Role

We are looking for a hands-on Computer Vision Engineer to work on the models that power Ripik's industrial AI platform. You will own CV problems end-to-end — from data strategy and annotation to model development, edge deployment, and production monitoring — for some of the most complex vision problems in heavy industry.

Key Responsibilities

• Own computer vision problems end-to-end — from problem framing and data strategy through model development, edge deployment, and production monitoring — across Ripik's industrial portfolio (steel, cement, pharma, paints, and beyond).
• Build models for hard vision challenges — novel defect types, extreme class imbalance, multi-camera fusion, low-light / high-noise factory environments, and real-time inference on constrained edge hardware.
• Stay at the cutting edge of CV research and rapidly evaluate and adopt new models and techniques — YOLO26, SAM 3, Vision Transformers (DINOv2, Swin), Grounding DINO, RF-DETR, zero-shot / open-vocabulary detection (YOLO-World, CLIP) — translating papers into production value.
• Follow and contribute to engineering standards for the vision stack — model training pipelines, data versioning (DVC), annotation workflows (CVAT, Roboflow, Label Studio), experiment tracking (W&B, MLflow), edge export formats (TensorRT, ONNX, OpenVINO), and CI/CD for model updates.
• Drive inference optimisation — quantisation (INT8 / FP16, GPTQ), pruning, knowledge distillation, and batching strategies — to meet latency and cost targets across NVIDIA Jetson, industrial PCs, and cloud GPU instances.
• Debug production issues on live customer deployments — trace performance drops, root-cause failure modes, and ship fixes with the right guardrails.
• Champion a data-centric AI approach — invest in annotation quality, active learning, synthetic data generation, and feedback loops from production rather than only chasing bigger models.
• Build robust evaluation frameworks — domain-specific metrics, A/B testing against production baselines, and systematic failure-mode analysis to ensure models deliver real business impact.
• Partner cross-functionally with product, field engineering, operations, and leadership — translate business problems into well-scoped modelling projects and communicate results clearly.

What you bring

• Bachelor's in Computer Science, AI/ML, Electrical Engineering, or a related field.
• 1–3 years of hands-on experience in computer vision — with a strong track record of taking models from research / prototyping through to production deployment.
• Deep proficiency in Python and PyTorch; strong working knowledge of OpenCV, Albumentations, and image / video processing fundamentals.
• Demonstrated expertise across multiple CV tasks — object detection, instance / semantic / panoptic segmentation, anomaly detection, pose estimation, or tracking.
• Hands-on experience with modern model families — YOLO (v8 / v11 / v26), transformer-based detectors (RT-DETR, DETR, RF-DETR), segmentation models (SAM / SAM 2), and CNN backbones (ResNet, EfficientNet, ConvNeXt, Vision Transformers).
• Production experience deploying models to edge or on-prem hardware using TensorRT, ONNX Runtime, or OpenVINO; comfort with Docker, Kubernetes, and at least one cloud platform (AWS / Azure / GCP).
• Strong first-principles problem-solving — comfortable navigating novel, unstructured problems where no playbook exists.
• Experience in a high-growth start-up or similarly fast-paced environment.
• Excellent communication — able to distil complex technical concepts for non-technical stakeholders, write clear documentation, and present results to leadership and customers.

Good to Have

• Experience with industrial or manufacturing domains — understanding of factory-floor constraints, camera setups, lighting variability, and integration with PLCs / SCADA systems.
• Familiarity with zero-shot and open-vocabulary detection (Grounding DINO, YOLO-World, CLIP) and foundation models (DINOv2, SAM 3, Florence) for data-efficient learning.
• Exposure to vision–language models (GPT-4o vision, Gemini, LLaVA) for combining visual inspection with natural-language reporting or operator copilots.
• Knowledge of 3D vision, depth estimation, point-cloud processing, or multi-camera calibration for volumetric industrial inspection.
• Experience with multi-object tracking (ByteTrack, BoT-SORT) and video analytics pipelines for continuous production-line monitoring.
• Contributions to open-source CV projects, publications in top-tier venues (CVPR, ECCV, ICCV, NeurIPS), or strong Kaggle competition results.