I’m Awais Asghar, an AI and Computer Vision Engineer building production-ready vision, LLM, robotics, and edge AI solutions—from research to deployment.
Awais Asghar
AI & Computer Vision Engineer
4+
Years in AI
30+
Projects
I’m an AI and Computer Vision Engineer with 4+ years of experience developing systems that move from research and model training to real-time deployment. My core work includes object detection, tracking, segmentation, crowd analytics, autonomous systems, and optimized inference for resource-constrained hardware.
At Cavalier Group, I contribute to production AI for UAVs, surveillance, payload systems, and edge devices. I work across the complete engineering pipeline—from dataset preparation and model evaluation to ONNX and TensorRT optimization and deployment on NVIDIA Jetson, Raspberry Pi, and Hailo hardware. Sensitive defense implementation details are protected by NDA.
Beyond computer vision and robotics, I build LLM, RAG, agentic, OCR, and document-intelligence applications using tools such as LangChain, LangGraph, Pinecone, ChromaDB, GPT, Gemini, and Llama. I’m most interested in roles where accuracy, deployment efficiency, and dependable real-world performance matter as much as the model itself.
Developing advanced AI models for high-precision object detection and tracking in dynamic environments.
Implementing AI models on edge devices like Raspberry Pi 5 and Jetson Nano for real-time applications.
Designing algorithms for optimizing performance in challenging conditions, such as variable wind dynamics and altitude.
My professional journey in AI engineering and computational science.
Blitzkrieg Defense Solution, Cavalier Group
30+ projects spanning computer vision, autonomous drones, defense AI, LLM applications, robotics, and edge deployment.
Defense work protected by NDA
Source code, operational footage, client data, system architecture, and sensitive performance specifications cannot be published. The summaries below describe only non-confidential aspects of my responsibilities and technical experience.
Research-driven density estimation for counting crowds in high-altitude drone imagery, with inference and validation-performance evaluation.
Custom two-class YOLO detector trained on about 7,000 drone and VisDrone images, with tracking and deployment optimization.
Real-time long-range object detection, recognition, and multi-object tracking integrated with NVIDIA Jetson hardware.
Predicts an accurate payload release point using altitude, speed, wind conditions, drone dynamics, and ballistic modeling.
Autonomous, real-time target locking and object-following system designed for drone-mounted cameras.
Autonomous drone navigation that combines computer vision, target tracking, and flight control for reliable subject following.
Real-time person and vehicle detection with tracking for continuous aerial and fixed-camera surveillance.
Video analytics pipeline for object detection, counting, tracking, and automated event monitoring.
Industry platforms that demonstrate production capabilities aligned with my work in venue analytics, real-time computer vision, and edge AI deployment.
Official UAV division reference covering autonomous VTOL platforms, ISTAR, target tracking, long-range surveillance, airframe and avionics development, edge computing, and subsystem integration.
A production example of people counting, dwell-time measurement, crowd-flow analytics, and automated safety alerts for high-footfall venues.
Technical documentation for production AI inference, crowd estimation, tracking, VMS integrations, and deployment across Jetson, Hailo, GPUs, and cloud infrastructure.
Product names, imagery, and documentation belong to their respective owners. Links are provided for technical and industry context.
My research focuses on innovative approaches to machine learning and AI systems.
A decentralized machine learning framework that implements Federated Learning to enable collaborative model training while preserving data privacy. The system distributes training data across multiple edge devices/clients.
Collaborators:
A recruiter-friendly view of my AI engineering stack, from model development to production deployment.
My academic background and professional certifications.
Focused on artificial intelligence, signal processing, embedded systems, control systems, and electrical engineering fundamentals.
Introduction to Power Electronics
Interested in collaborating on AI projects or have questions about my work? Feel free to reach out.