Python CV Engineer for Image Enhancement/Processing
Budget: $30 – $250 USD
I’m building a production-grade image-enhancement pipeline and want an AI engineer who lives and breathes Python and computer vision. The goal is to take raw mobile photos and automatically clean noise, sharpen details, fix color and convert them to 3d printable files. You’ll choose or design the model, train or fine-tune it, then wrap everything in a lightweight API that my mobile team can call in real time (on-device when feasible, cloud fallback when not).
You should be completely comfortable with OpenCV plus deep-learning frameworks such as PyTorch or TensorFlow, and you know the trade-offs between traditional filters, GAN-based approaches, and modern super-resolution networks. Experience packaging models for CoreML, TensorFlow Lite or similar mobile runtimes will set you apart.
I’m most interested in seeing what you’ve already shipped, so please include past work that proves you can take an image-processing idea from notebook to app store.
Deliverables
• Clean, well-commented Python codebase and requirements.txt
• Trained weights with reproducible training scripts
• Mobile-ready wrapper (REST, gRPC, CoreML, or TFLite)
• README with run instructions and basic quality benchmarks (PSNR/SSIM)
Acceptance criteria: pipeline meets agreed-upon quality metrics on our validation set, passes automated tests and runs smoothly in a demo app on both iOS and Android. If you have the skills and past results to match, let’s create stunning images together.
You should be completely comfortable with OpenCV plus deep-learning frameworks such as PyTorch or TensorFlow, and you know the trade-offs between traditional filters, GAN-based approaches, and modern super-resolution networks. Experience packaging models for CoreML, TensorFlow Lite or similar mobile runtimes will set you apart.
I’m most interested in seeing what you’ve already shipped, so please include past work that proves you can take an image-processing idea from notebook to app store.
Deliverables
• Clean, well-commented Python codebase and requirements.txt
• Trained weights with reproducible training scripts
• Mobile-ready wrapper (REST, gRPC, CoreML, or TFLite)
• README with run instructions and basic quality benchmarks (PSNR/SSIM)
Acceptance criteria: pipeline meets agreed-upon quality metrics on our validation set, passes automated tests and runs smoothly in a demo app on both iOS and Android. If you have the skills and past results to match, let’s create stunning images together.