Job Description: Junior Computer Vision Engineer (YOLO, Edge & GPU Deployment)

Role Overview

We are looking for a Junior Computer Vision Engineer (2-4 years experience) to build and deploy real-time vision models using the YOLO family. The role focuses on hands-on model development, basic distillation techniques, and deployment on CPU and NVIDIA GPU environments, with optional exposure to cloud platforms.

This is an execution-heavy role suited for someone who enjoys working close to systems, models, and performance.

Key Responsibilities

Model Development

- Develop and fine-tune YOLO-based models (e.g., YOLOv5, YOLOv8) for object detection and basic segmentation tasks
- Assist in implementing knowledge distillation (teacherstudent models) to reduce model size and improve inference speed
- Work on dataset preparation, annotation validation, and augmentation

Model Optimization

- Apply basic optimization techniques: quantization (INT8 / FP16) and model pruning (guided)
- Convert models to deployable formats using ONNX and TensorRT

Deployment & Inference

- Deploy models on CPU-based environments (edge devices, servers) and NVIDIA GPUs (e.g., T4, L4, Jetson)
- Assist in building inference pipelines using OpenCV and PyTorch
- Support performance tuning for latency and throughput

Cloud (Optional Exposure)

- Support deployment pipelines on cloud platforms such as AWS, Google Cloud, or Azure
- Learn containerization basics (Docker)



and model serving concepts

Required Qualifications

- Bachelors degree in Computer Science, AI, Electronics, or related field
- 2-4 years of experience in computer vision or deep learning
- Robust fundamentals in Python programming and deep learning basics (CNNs, training workflows)
- Hands-on experience with PyTorch or TensorFlow and basic OpenCV usage
- Familiarity with Linux environments

Preferred Qualifications

- Exposure to YOLO family models (training or inference)
- Understanding of model compression / distillation concepts
- Basic understanding of GPU vs CPU inference trade-offs
- Experience with Docker and REST APIs or simple backend integration
- Exposure to edge devices like NVIDIA Jetson is a plus

Key Skills

- Problem-solving and debugging
- Ability to optimize for performance (latency, memory)
- Basic understanding of real-time systems
- Willingness to learn deployment and systems aspects

What You Will Gain

- Hands-on experience with production-grade computer vision systems
- Exposure to edge + GPU deployment at scale
- Learning opportunity in model optimization and real-time inference systems

Nice-to-Have Project Experience

- Real-time video analytics (people detection, PPE detection, etc.)
- Edge deployment projects (Jetson / CPU inference pipelines)
- Projects involving model compression or speed optimization

📌 Computer Vision Engineer (Hyderabad)
🏢 Inumellas Consultancy Services
📍 Hyderabad