Software Engineer, AI/ML - Computer Vision

About the role

Devsinc is hiring a Software Engineer with 2+ years of experience and a strong foundation in Machine Learning and Computer Vision for a health tech project in Islamabad. The project builds real-time AI assistance for clinicians performing ultrasound-guided endoscopic procedures, analyzing live ultrasound video and highlighting anatomy, suspected pathology and instruments frame by frame. The role covers the segmentation model lifecycle, from data preparation and training to evaluation, deployment, retraining and direct client collaboration.

Key responsibilities

• Train and improve multi-label medical image segmentation models using PyTorch, with ultrasound video and clinician-annotated data
• Run experiments on class imbalance, overlapping anatomy, small or thin targets and temporal context
• Build evaluation pipelines with per-label, frame-level and event-level metrics, validated against patient-level test splits and clinician review
• Optimize inference for real-time video with GPU latency in mind, using mixed precision, model compilation, batching and frame stride
• Package and deploy models with Docker and AWS-based infrastructure, and contribute reproducible training and retraining pipelines with experiment tracking, checkpointing, CI and automated testing
• Collaborate with clinicians, annotators, infrastructure engineers and external ML partners, and take part in client calls, demos and requirement discussions
• Explain technical decisions, results, risks and trade-offs clearly to technical and non-technical stakeholders

Required profile

• Bachelor's degree in Computer Science or Artificial Intelligence
• 2+ years of software engineering, machine learning or computer vision experience
• Strong Python skills and hands-on PyTorch experience
• Experience training image segmentation or computer vision models
• Good communication skills and confidence in client-facing technical discussions

Required skills

• PyTorch, image segmentation and computer vision
• AWS, Docker, cloud GPU environments and object storage
• Production ML inference and performance optimization
• Testing, version control, reproducible configurations and CI

Nice to have: medical imaging or segmentation experience, real-time computer vision, multi-GPU training, and prior client-facing work. Benefits include Provident Fund, medical inpatient and outpatient facilities, paid overtime, subsidized meals, gym facility, sports allowance, interest-free loan facility, advance salary options and employee activities.