Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I design and ship practical AI—combining robust backends with models that run efficiently on the edge. Recent work includes a bilingual RAG assistant on Azure (ADF ingestion, OpenAI embeddings, Azure AI Search, Flask/Docker, similarity gating) and embedded CV: TinyML facial recognition on ESP32-CAM, depth-aware obstacle avoidance, and indoor navigation with QR landmarks, all under tight latency/memory budgets. I’m fluent in Python/C++, OpenCV, PyTorch/TensorFlow, Hugging Face, and TFLite-Micro/TensorRT; comfortable with SQL/PostGIS, REST APIs, Docker, GitHub Actions, and observability. I value clean architecture, measurable outcomes, and shipping small, reliable increments. My goal: build disciplined, production-ready Edge-AI that makes real-world systems smarter and safer.…I design and ship practical AI—combining robust backends with models that run efficiently on the edge. Recent work includes a bilingual RAG assistant on Azure (ADF ingestion, OpenAI embeddings, Azure AI Search, Flask/Docker, similarity gating) and embedded CV: TinyML facial recognition on ESP32-CAM, depth-aware obstacle avoidance, and indoor navigation with QR landmarks, all under tight latency/memory budgets. I’m fluent in Python/C++, OpenCV, PyTorch/TensorFlow, Hugging Face, and TFLite-Micro/TensorRT; comfortable with SQL/PostGIS, REST APIs, Docker, GitHub Actions, and observability. I value clean architecture, measurable outcomes, and shipping small, reliable increments. My goal: build disciplined, production-ready Edge-AI that makes real-world systems smarter and safer.

Aanal Kamleshkumar Patel

AI Developer, AI Engineer, AI Prompt Engineer, +3





I design and ship practical AI—combining robust backends with models that run efficiently on the edge. Recent work includes a bilingual RAG assistant on Azure (ADF ingestion, OpenAI embeddings, Azure AI Search, Flask/Docker, similarity gating) and embedded CV: TinyML facial recognition on ESP32-CAM, depth-aware obstacle avoidance, and indoor navigation with QR landmarks, all under tight latency/memory budgets. I’m fluent in Python/C++, OpenCV, PyTorch/TensorFlow, Hugging Face, and TFLite-Micro/TensorRT; comfortable with SQL/PostGIS, REST APIs, Docker, GitHub Actions, and observability. I value clean architecture, measurable outcomes, and shipping small, reliable increments. My goal: build disciplined, production-ready Edge-AI that makes real-world systems smarter and safer.…I design and ship practical AI—combining robust backends with models that run efficiently on the edge. Recent work includes a bilingual RAG assistant on Azure (ADF ingestion, OpenAI embeddings, Azure AI Search, Flask/Docker, similarity gating) and embedded CV: TinyML facial recognition on ESP32-CAM, depth-aware obstacle avoidance, and indoor navigation with QR landmarks, all under tight latency/memory budgets. I’m fluent in Python/C++, OpenCV, PyTorch/TensorFlow, Hugging Face, and TFLite-Micro/TensorRT; comfortable with SQL/PostGIS, REST APIs, Docker, GitHub Actions, and observability. I value clean architecture, measurable outcomes, and shipping small, reliable increments. My goal: build disciplined, production-ready Edge-AI that makes real-world systems smarter and safer.

Available to hire

I design and ship practical AI—combining robust backends with models that run efficiently on the edge. Recent work includes a bilingual RAG assistant on Azure (ADF ingestion, OpenAI embeddings, Azure AI Search, Flask/Docker, similarity gating) and embedded CV: TinyML facial recognition on ESP32-CAM, depth-aware obstacle avoidance, and indoor navigation with QR landmarks, all under tight latency/memory budgets. I’m fluent in Python/C++, OpenCV, PyTorch/TensorFlow, Hugging Face, and TFLite-Micro/TensorRT; comfortable with SQL/PostGIS, REST APIs, Docker, GitHub Actions, and observability. I value clean architecture, measurable outcomes, and shipping small, reliable increments. My goal: build disciplined, production-ready Edge-AI that makes real-world systems smarter and safer.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Intermediate

Intermediate

Language

English

Fluent

Work Experience

Research Assistant at Lambton Research and Innovation

June 1, 2025 - Present

Designed and deployed a bilingual FAQ AI chatbot using a Retrieval-Augmented Generation architecture for Tech-Access Canada’s iVisit program. Implemented automated PDF ingestion, text chunking, and embedding generation workflows using Azure Data Factory and OpenAI embedding API. Built a Flask-based backend, containerized with Docker, and deployed to Azure Container Instances with RESTful API endpoints for query processing. Integrated OpenAI GPT models to generate short and detailed responses with high accuracy via strict similarity gating and contextual retrieval. Used Azure Cognitive Services Language Understanding (CLU) for language detection and intent normalization, improving overall system accuracy to 95%. Led development of a computer-vision-based inventory system using Post-Training Quantization and Knowledge Distillation to reduce model footprint for ARM-based embedded hardware.

Associate AI Developer at Credwise Financial Inc.

May 1, 2024 - December 31, 2024

Developed a location-based recommender API to optimize user spend by identifying high-reward options. Implemented a multi-modal API integrating OCR-powered vision processing, spend-behavior analysis, and budget tracking. Authored comprehensive technical documentation and a developer-friendly Wiki to streamline onboarding. Built a back-end image-generation API creating dynamic, category-based advertising banners for campaigns. Implemented a merchant locator API to retrieve and structure merchant data by category and city, enhancing location-based services.

R&D Engineer at Prayosha Food Services Pvt Ltd

March 1, 2022 - July 31, 2023

Architected a real-time obstacle-avoidance pipeline for autonomous service robots using Intel RealSense depth cameras and Lightweight CNNs (MobileNetV3). Achieved 20% reduction in navigation latency and improved localization precision in dynamic environments. Developed and deployed a lightweight facial recognition system on an ESP32-CAM using TinyML for real-time identity verification on resource-constrained edge devices. Engineered secure OTA firmware updates on embedded micro-controllers. Built a hybrid robot navigation system by integrating IR cameras, ceiling-mounted QR code tracking, and depth cameras for precise indoor localization and obstacle avoidance. Led core development of robot control code, managed Git version control with structured branching, and maintained regular updates for seamless collaboration.

AI ML Intern at GIRO FinTech

July 1, 2021 - October 31, 2021

Developed a hand gesture control system for gaming applications, fine-tuned ML models, and designed hand-tracking and anti-cheat ML models to improve security and fairness on a gaming platform.

Volunteer Tech Lead at Yiddishland, California

January 1, 2025 - August 1, 2025

Managed task distribution and team coordination, improving workflow efficiency by 20% through effective delegation and prioritization. Maintained regular communication with stakeholders, ensuring 100% adherence to project timelines and deliverables.

Research Team Member at Robotics Club, Vadodara Institute of Engineering

March 1, 2022 - July 31, 2023

Contributed to embedded programming, machine-learning applications in embedded systems, and hardware development. Gained hands-on experience with C++, C, Arduino programming, and IoT technologies. Explored foundational communication protocols such as I2C, SPI, and UART through hands-on projects involving microcontrollers and embedded systems.