Computer Vision Engineer (YOLO + OCR Integration for Live Poker Tables)

Ideally, we are looking for a developer who already has a working or partially ready solution for poker table recognition and can quickly integrate or adapt it to our requirements.

 

We are looking for a skilled Computer Vision / ML Engineer to integrate ready-made open-source solutions (YOLO + PaddleOCR) into our real-time poker table parser.

Your task:
Take an existing repository such as yolo11-poker-hand-detection-and-analysis or similar, and adapt it to our poker client UI.
The system must detect cards, stacks, pot size, and player actions from a video stream or screen capture in real-time (1โ€“6 tables simultaneously) and output structured JSON events with minimal latency (<0.5s).

โš™๏ธ Responsibilities

  • Integrate YOLOv10 / YOLO11 or RT-DETR model for detecting poker elements.
  • Use PaddleOCR (custom dictionary: cards, suits, numbers) to extract text (stacks, pot, etc).
  • Adapt detection zones and templates to our poker room UI (multi-table layout).
  • Implement multi-threaded pipeline for 1โ€“6 tables in parallel.
  • Output standardized JSON events (or WebSocket stream) per detected action.
  • Ensure high accuracy and low latency (target: <0.5s delay).

๐Ÿงฉ Requirements

  • Experience with YOLO (v8โ€“v11), ONNX / TensorRT, PaddleOCR, OpenCV.
  • Strong Python or C++ skills.
  • Understanding of real-time computer vision, frame capture, and performance optimization.
  • Ability to work independently and deliver stable prototypes quickly.
  • Experience with gaming/video analytics or OCR tasks will be a plus.

Required languages

Ukrainian Native
Published 5 October
52 views
ยท
2 applications
100% read
To apply for this and other jobs on Djinni login or signup.
Loading...