z.systems is a fast‑growing, VC‑backed technology company that is revolutionizing traditional retail across Africa. By creating a full‑stack B2B2C digital ecosystem, we connect independent grocery stores with brands, customers, and a loyalty engine that transforms transactional relationships into long‑term brand affinity. With over 20,000 retailers onboarded and 900,000 orders delivered, we are building the future operating model for informal retail, offering dynamic promotions, gamified loyalty programs, and buy‑now‑pay‑later features.

As the ML Technical Lead – Computer Vision & Shelf Recognition, you will be at the forefront of this transformation. Your primary responsibilities include:

• Team Leadership: Lead a small CV/ML engineering team, mentor junior engineers, and foster a culture of continuous improvement and innovation.
• Roadmap & Architecture: Define the technical roadmap for detection and recognition models, evaluate cutting‑edge architectures such as YOLOv8, DINOv3, and ViT backbones, and decide on OBB vs. HBB strategies for small object detection.
• Pipeline Design: Build end‑to‑end training pipelines that incorporate real and synthetic datasets, data augmentation, annotation quality control, and tiling for small objects.
• Distributed Training: Implement multi‑GPU distributed training on AWS, ensuring scalability and efficiency.
• CI/CD & Deployment: Establish CI/CD workflows for model training, versioning, deployment, and A/B testing, integrating with our product and operations teams.
• Collaboration: Work closely with product managers, operations, and retail partners to deploy models into retailer‑facing applications, ensuring seamless integration and performance.

We expect you to bring:

• 5+ years of experience in computer vision and deep learning, with at least 2 years in a lead role.
• Proficiency in PyTorch, Ultralytics YOLO, and distributed training on AWS.
• Expertise in synthetic data generation, annotation pipelines, and handling small object detection challenges.
• Strong communication skills and the ability to lead cross‑functional teams.

Bonus skills include:

• Experience with DINO/DINOv2/v3, ViT‑based backbones, and on‑device optimization (TensorRT, TFLite).
• Background in retail tech, OCR, or dense shelf detection.

The recruitment process involves CV pre‑screening, an AI interview, and a technical and fit interview with the Zsystem team. We value innovation, collaboration, and a passion for transforming retail through technology.

Join us to lead cutting‑edge computer vision solutions that empower retailers across Africa, driving efficiency, visibility, and customer engagement at scale.