Rishabh Parmar
About Candidate
I am currently working as Jr. AI/ML Engineer at WebOccult with a strong
focus on Computer Vision and Edge AI, experienced in designing, deploying,
and optimizing real-time AI applications for image classification, object detection,
segmentation, and OCR on embedded platforms. Proficient in deploying AI models
on GPUs (Jetson), NPUs, and AI accelerators, delivering efficient, low-latency
solutions through hardware-aware optimization and camera integration. Driven
by curiosity and a builder’s mindset, I enjoy turning ideas into production-ready
systems, exploring technologies at a deeper level, and continuously learning to
solve challenging real-world problems.
Location
Education
CPI : 8.55 – Percentage: 80.5%
Work & Experience
• Designed and optimized end-to-end Computer Vision inference pipelines for
real-time Edge AI.
• Built pipelines for classification, detection, segmentation and OCR using YOLO
and PaddleOCR.
• Optimized latency, FPS, memory utilization and runtime performance.
• Integrated USB, IP (RTSP), CSI and MIPI cameras with on-site deployment
and operations experience.
• Worked on NXP i.MX 8M Plus, Jetson Orin Nano/AGX/Thor, Rockchip,
Axelera, MediaTek Arbor SBC, Amobile G700 and Raspberry Pi IMX500.
• Built Skittle, mobile and container detection models with CPU-based OCR.
• Developed AI demos for CES, Embedded World, EuroShop, and other global
expos.
• Performed on-site deployment, integration, valid
• Worked with OpenCV, NLP, dataset annotation, and AI inference pipeline
development on DeepX and Deeper-i NPUs.
• Developed and deployed an end-to-end Edge AI demo for Japan IT Week in
collaboration with Deeper-i.
• Applied core AI/ML concepts to develop a Driver Behavior Monitoring System
using real-time image processing, while gaining hands-on experience with
Python libraries, web scraping, neural networks, and cloud computing.