Browse categories
Explore
Fiverr Pro
English
$
USD
I will optimize and deploy your ai model for edge devices
Rukon Uddin
About this gig
Is your AI model too slow, too heavy, or stuck running only on a GPU server? I help you take a trained model and make it actually work in the real world on a phone, a Raspberry Pi, an NVIDIA Jetson, or a custom NPU chip.
What I do:
- Convert your model to ONNX, TensorFlow Lite, CoreML, or vendor-specific NPU formats (including HiSilicon ATC)
- Quantize and prune for faster inference and a smaller footprint, with minimal accuracy loss
- Deploy and benchmark on your target device latency, memory, and accuracy trade-offs, clearly reported
- Debug real deployment issues: driver mismatches, pre/post-processing bugs, memory crashes
Why work with me: I'm a computer vision engineer with 5 years of experience building production ML systems including an airport indoor-navigation system I'm currently building for visually impaired travelers plus a Kaggle Computer Vision Silver medal.
Send me your model type and target hardware before you order, and I'll tell you exactly what's achievable and the fastest way to get there.
Get to know Rukon Uddin
Rukon Uddin
AI Engineer
- FromBangladesh
- Member sinceJul 2026
- Avg. response time1 hour
Languages
English, Arabic, Spanish
AI/ML engineer with 5 years building the harder side of applied AI: LiDAR SLAM, sensor fusion, and vision-language model deployment on edge hardware.
I built a LiDAR-based 3D mapping pipeline (FAST-LIVO2) fusing LiDAR, IMU, and camera data, plus a vision-based navigation system for visually impaired users running live at Hong Kong Airport on minimal VRAM. I specialize in making it run fast and reliably on real hardware.
Top 4% finisher, TensorFlow-sponsored Kaggle competition.
Robotics, sensor fusion, real-time systems, or edge deployment — let's talk.
