Über
This role requires strong expertise in deep learning (PyTorch), computer vision, backend API development (FastAPI), containerization (Docker), and AWS deployment. You will be responsible for building a scalable, high-performance inference system from model development through cloud deployment.
This is a high-impact project with potential for long-term collaboration.
Project Scope:
We are building an object detection solution that:
- Uses state-of-the-art PyTorch-based detection models (YOLO, Faster R-CNN, etc.)
- Performs real-time inference on images and video streams
- Exposes prediction endpoints via FastAPI
- Uses OpenCV for preprocessing and video handling
- Is fully containerized using Docker
- Deploys on AWS infrastructure
- Supports GPU-based inference
- Is optimized for production performance and scalability
Key Responsibilities:
- Design and implement object detection pipelines using PyTorch
- Optimize models for real-time performance (CPU/GPU)
- Build clean, scalable APIs using FastAPI
- Implement image/video processing with OpenCV
- Containerize services with Docker
- Deploy and manage services on AWS (EC2, ECS, ECR, S3, IAM, etc.)
- Improve inference throughput and latency
- Ensure clean architecture and maintainable codebase
- Provide documentation and deployment guidelines
Required Qualifications:
- 5+ years experience in Machine Learning / Deep Learning
- Strong hands-on experience with PyTorch
- Proven experience building and deploying object detection systems
- Experience optimizing models for production environments
- Strong Python engineering skills
- Experience with FastAPI (production-level systems)
- Experience with Docker
- Experience deploying ML systems on AWS
- Experience working with GPU environments (CUDA)
Preferred Qualifications:
- Model optimization experience (quantization, pruning, TensorRT)
- Experience with streaming pipelines (RTSP/WebRTC)
- MLOps experience
- CI/CD and infrastructure-as-code experience
- Experience with large-scale inference systems
Deliverables:
- Production-ready object detection API
- Dockerized and documented deployment
- AWS-based scalable infrastructure
- Optimized inference pipeline (low latency)
- Clean GitHub repository with documentation
To Apply, Please Include:
1. Links to relevant ML/computer vision projects
2. Details of your experience with PyTorch object detection
3. Example of AWS ML deployment you've worked on
4. Estimated timeline for MVP delivery
5. Your rate and availability
Contract duration of 1 to 3 months.
Mandatory skills: Machine Learning, Python, Deep Learning, Data Science, TensorFlow, Computer Vision
Sprachkenntnisse
- English
Hinweis für Nutzer
Dieses Stellenangebot stammt von einer Partnerplattform von TieTalent. Klicken Sie auf „Jetzt Bewerben“, um Ihre Bewerbung direkt auf deren Website einzureichen.