Open role
Apply for this roleComputer Vision Engineer
Key requirementHands-on experience with GStreamer or NVIDIA DeepStream is essential for this role.
What you will do
- Build computer vision models that process and analyze live video in real time.
- Work with RTSP and IP surveillance cameras, and handle frame capture and decoding at scale.
- Design scalable, deployable microservices.
- Integrate with databases to store, analyze and track detection data.
- Write high-performance code using multi-threading, multi-processing and async I/O.
- Deploy solutions on the cloud (AWS, Azure and others) and on edge devices.
Nice to have
- Messaging systems: Kafka, RabbitMQ or Redis Streams.
- Containerization with Docker.
- Vision-language models (VLMs) and a habit of following the latest AI research.
What you need
- Bachelor’s or Master’s degree in Computer Science or a related field.
- At least 1 year of professional computer vision experience.
- Strong experience with OpenCV, MediaPipe, YOLO / Ultralytics and DeepSORT.
- Hands-on experience with GStreamer or NVIDIA DeepStream.
- Databases: PostgreSQL and MongoDB.
- Building APIs with FastAPI or Flask.
- Model quantization and optimization to reduce model size and latency.
- Edge devices such as NVIDIA Jetson, Raspberry Pi or Intel NCS.
- Comfortable with Linux system administration, Git and GitHub.
- Familiarity with generative AI models.
Our stack
PythonOpenCVYOLOUltralyticsDeepSORTMediaPipeGStreamerDeepStreamFastAPIPostgreSQLMongoDBKafkaRedisDockerJetsonAWSAzure