DontApply.Club
Login Register

Computer Vision Engineer

Aman Global — Kazakhstan · Posted ~4 hours ago

Mid Part-time Remote

Skills

Computer Vision Machine Learning Model Development Production Deployment ML Pipeline Video Analytics CCTV

🔓 Log in to save this job, tailor your resume & track your apply process — 7 days free, no card needed.

Log in to add to target list

Summary ✨ AI‑Generated

A startup specializing in computer vision-powered video analytics for surveillance systems is opening a new product direction focused on the academic sector. The system is already live and scaling in international markets. The new Computer Vision Engineer will lead a small team (including supportive interns) and own the full ML pipeline — from model development through production deployment. This is a hands-on role with real technical ownership, not a research-only position. The team is distributed across two major cities and works remotely on a part-time basis (~20 hours per week). Candidates will collaborate closely with the CPO, CTO, backend engineers, and another CV engineer working on the existing product line.

Highlights

Remote part-time role (~20 hrs/week) with real ownership of the ML pipeline. Opportunity to lead a new product direction within an established computer vision system. Small, technically strong team with supportive interns for data collection and testing. Live production system with growing international presence.

Description

Format: Remote, Part-time (~20 hrs/week; teams collaborate across Almaty and Astana) About the Company AMAN Ltd. is a startup building computer vision-powered video analytics for CCTV and surveillance systems. Our current product line covers security and private facilities. We're now opening a new direction: the academic sector, focused on integrity and engagement — and you'll lead it. This isn't a research exercise — it's a real production system already live, currently growing in UAE and EU markets. You'll be joining a small, technically strong team: a CPO, a hands-on CTO with 10 years of industry experience, a Backend Engineer, a CV Engineer (working on the existing product), and 2 interns ready to support you with data collection, annotation, and testing. As the CV Engineer leading this new direction, you'll have real ownership over the ML pipeline that powers it — from model development through to production deployment — and the chance to shape how the system scales into this new domain. What You'll Do Develop and improve human action recognition models on videoWork with video streams (RTSP, IP cameras, NVR)Process video in real time (10–25 FPS)Prepare and clean datasets, labeling, augmentationsTrain and fine-tune models (PyTorch)Optimize inference (GPU/CPU)Reduce false positives / false negativesIntegrate the model into a backend service (API / queues / microservices)Generate event clips (pre/post buffer) Our Stack Python 3PyTorchOpenCVYOLO / action recognition RTSP video streams DockerLinuxREST API Requirements Required: Strong Python skillsExperience with PyTorch or TensorFlowUnderstanding of computer vision (detection / tracking / classification)Experience working with OpenCVUnderstanding of how neural networks are trained (loss, overfitting, augmentation)Strong plus: Experience with RTSP/IP cameras YOLO, DeepSORT, action recognition TensorRT / ONNX / inference acceleration GPU optimization Contacts to apply Email: Akerke.bexultan@aman.org.kz Telegram: @akerke_bx