Computer Vision Lead

Role Overview:

Computer Vision Lead will head the vision and multimodal AI charter behind our next-generation sports AI products. This is a hands-on leadership role: you will own the technical direction of our real-time sports understanding stack, set the architecture and research agenda, and build and mentor a high-performing team of Computer Vision and AI Engineers. You will be accountable for taking models from research to reliable, low-latency production systems experienced by millions of fans.

Key Responsibility Areas:

1. Technical Leadership & Team Building

●     Own the technical vision, architecture, and roadmap for Computer Vision and multimodal AI across the product portfolio.

●     Lead, mentor, and grow a team of Computer Vision and AI Engineers; drive hiring, onboarding, and capability development.

●     Set engineering standards for experimentation, code quality, documentation, reproducibility, and model governance.

●     Translate product goals into technical milestones, effort estimates, and delivery plans; own execution end to end.

●     Make build-vs-buy and architecture trade-off decisions across model, infrastructure, and vendor choices.

2. AI Product Development & Language Intelligence

●     Define the strategy for AI-powered language products that enhance sports content creation and fan engagement.

●     Lead development of automated live commentary systems using Large Language Models (LLMs), multimodal AI, and speech technologies.

●     Architect intelligent pipelines that fuse vision, audio, and contextual match data into real-time insights and narratives.

●     Evaluate and adopt emerging AI architectures to keep product capabilities ahead of the market.

3. Computer Vision & Model Development

●     Direct the design, training, and optimization of computer vision models for live sports analytics.

●     Guide algorithm development for image and video understanding, including player, ball, and object tracking.

●     Set architecture direction across CNNs, Vision Transformers (ViTs), YOLO, Faster R-CNN, Mask R-CNN, and vision-language models.

●     Oversee object detection, image classification, segmentation, pose estimation, OCR, facial recognition, and event detection workstreams.

●     Establish data strategy: annotation pipelines, augmentation, dataset quality, and data engineering best practices.

●     Define evaluation frameworks and benchmarks for accuracy, latency, and robustness; drive continuous improvement against them.

4. Deployment, Performance Optimization & Production Engineering

●     Own the deployment architecture for models across cloud, edge devices, and GPU-enabled environments.

●     Architect scalable inference pipelines with high throughput and low latency for real-time applications.

●     Drive deployment practices using Docker, ONNX, TensorRT, FastAPI, Kubernetes, and edge GPU platforms.

●     Ensure robust integration into production systems through APIs, microservices, and modern software engineering practices.

●     Establish monitoring, observability, and MLOps practices for production reliability and cost efficiency.

5. Innovation, Collaboration & Research

●     Partner with Product, Design, Data Science, and Software Engineering leadership to shape the product roadmap.

●     Lead applied research in Computer Vision, Multimodal AI, Generative AI, and Sports Analytics; identify what is worth productionizing.

●     Represent the team in cross-functional and leadership forums; communicate technical trade-offs to non-technical stakeholders.

●     Champion engineering excellence through architecture reviews, design reviews, and code reviews.

Ideal candidate will have:

●     8–12 years of hands-on experience in Computer Vision, Machine Learning, or Applied AI, including 2+ years leading or mentoring engineering teams.

●     B.E./B.Tech./M.Tech. in Computer Science, Artificial Intelligence, Electronics, Robotics, or a related discipline; PhD is a plus.

●     Proven track record of shipping computer vision systems to production at scale, ideally in real-time or low-latency environments.

●     Deep fundamentals in Machine Learning, Deep Learning, Computer Vision, NLP, and Multimodal AI.

●     Expertise in Python with strong software engineering and system design skills.

●     Deep hands-on experience with PyTorch and/or TensorFlow.

●     Strong command of transformer architectures and vision-language models.

●     Practical, production-grade experience across:

○     Object Detection

○     Pose Estimation

○     Player & Object Tracking

○     Image Classification

○     Image Segmentation

○     OCR

○     Automatic Speech Recognition (ASR)

●     Experience with model optimization and deployment (ONNX, TensorRT, quantization, edge GPU) and MLOps tooling.

●     Ability to balance research ambition with delivery discipline, and to communicate clearly across technical and business audiences.