█████╗ ██╗ ███████╗███╗ ██╗ ██████╗ ██╗███╗ ██╗███████╗███████╗██████╗
██╔══██╗██║ ██╔════╝████╗ ██║██╔════╝ ██║████╗ ██║██╔════╝██╔════╝██╔══██╗
███████║██║ █████╗ ██╔██╗ ██║██║ ███╗██║██╔██╗ ██║█████╗ █████╗ ██████╔╝
██╔══██║██║ ██╔══╝ ██║╚██╗██║██║ ██║██║██║╚██╗██║██╔══╝ ██╔══╝ ██╔══██╗
██║ ██║██║ ███████╗██║ ╚████║╚██████╔╝██║██║ ╚████║███████╗███████╗██║ ██║
╚═╝ ╚═╝╚═╝ ╚══════╝╚═╝ ╚═══╝ ╚═════╝ ╚═╝╚═╝ ╚═══╝╚══════╝╚══════╝╚═╝ ╚═╝
I am an AI Engineer focused on building production-grade intelligent systems — from LLM-powered agents and computer vision pipelines to explainable AI for high-stakes domains like medical imaging.
My work sits at the intersection of applied research and production engineering: I care about systems that are not just accurate, but interpretable, deployable, and reliable.
Currently building at LA Consulting Corp. — shipping Agentic AI systems and Computer Vision solutions.
Flagship research project (Explainable Lung Tumor Segmentation)
focus = [
"Agentic AI & Multi-Agent Systems",
"LLMs, RAG, and Tool-Augmented Reasoning",
"Computer Vision (Medical Imaging, Object Detection)",
"Explainable AI: SHAP, LIME",
"AI Automation & Orchestration (n8n)",
"Production ML Pipelines & Deployment",
]Core Languages
AI / LLMs / Agents
Deep Learning & Computer Vision
Explainability & Data
Deployment & Infrastructure
|
End-to-end CNN pipeline for lung tumor segmentation with LIME and SHAP-based explainability. Achieved 95% accuracy and 34% improvement in interpretability fidelity.
|
Production RAG system for semantic question answering over PDFs. Features a transparency layer that surfaces exactly which document sections informed each answer.
|
|
Content-based recommendation engine with integrated SHAP explainability, making each recommendation transparent and interpretable to end users. Deployed as a live Streamlit app.
|
Multi-step reasoning agents with tool-calling, persistent memory, and workflow orchestration — built for autonomous task execution in production environments.
|
I am open to roles in AI Engineering, LLM Systems, Agentic AI, and Applied ML Research — globally.
| kandhro62@gmail.com | |
| linkedin.com/in/abdus-samad-kandhro | |
| HuggingFace | huggingface.co/Samad80 |
| Live Demo | kandhro-ex-pdf-qa.hf.space |
"The goal is not to build AI that impresses — it is to build AI that works."