𝐀𝐈 𝐀𝐮𝐭𝐨𝐦𝐚𝐭𝐢𝐨𝐧 𝐄𝐧𝐠𝐢𝐧𝐞𝐞𝐫 | 𝐋𝐋𝐌 𝐀𝐠𝐞𝐧𝐭𝐬 | 𝐂𝐡𝐚𝐭𝐆𝐏𝐓 & 𝐖𝐨𝐫𝐤𝐟𝐥𝐨𝐰 𝐀𝐮𝐭𝐨𝐦𝐚𝐭𝐢𝐨𝐧 I build AI-powered systems that automate real business workflows using LLMs, agents, and modern automation tools. With 7+ years of experience in AI/ML and data engineering, I now specialize in designing and deploying production-ready AI solutions using ChatGPT, LangChain, and workflow automation platforms. I focus on delivering systems that don’t just “look good” they save time, reduce manual work, and scale operations. 🔧 What I Can Help You Build ✔ AI Agents for business automation (multi-step workflows) ✔ ChatGPT / OpenAI integrations (custom apps, APIs, tools) ✔ RAG systems (document Q&A, knowledge bases) ✔ AI chatbots for customer support & internal tools ✔ Workflow automation (Zapier, Make, APIs) ✔ NLP systems (classification, routing, extraction) ✔ End-to-end AI pipelines (from idea → production) ⚙️ Tech Stack OpenAI (GPT-4 / ChatGPT API) LangChain / LLM Agents Python (FastAPI, backend systems) Zapier / Make (automation) Vector DBs (Pinecone, FAISS) REST APIs & integrations 💼 Recent Work Includes Built AI chatbot handling real customer interactions Developed LLM-based routing system for NLP classification Automated workflows using ChatGPT + Zapier Deployed LangChain-based document AI system Integrated LLMs into production APIs 🎯 What You Can Expect Clean, scalable, production-ready solutions Strong communication and fast turnaround Focus on business outcomes (not just code)

I help companies build production-grade AI systems. With 9+ years of experience in AI/ML, I specialize in Document AI, OCR pipelines, Retrieval-Augmented Generation (RAG), and LLM-based agents used in real businesses across banking, healthcare, manufacturing, and e-commerce. If you’re struggling with unstructured data, PDFs, scanned documents, hallucinating LLMs, or brittle prototypes, I can help you take it to production. What I Build: 🔹 Document AI & OCR Systems • Invoice, contract, COA, compliance & regulatory documents • Custom OCR pipelines (AWS Textract, Azure Document Intelligence, PaddleOCR) • Layout-aware extraction using LayoutLM, custom CV + NLP hybrids • Entity extraction, tables, multi-page documents, noisy scans • Deployed at scale in banking & manufacturing environments 🔹 Retrieval-Augmented Generation (RAG) Systems • Enterprise RAG using Azure AI Search, Pinecone, Neo4j • Hybrid search (vector + keyword), metadata filtering, reranking • Query rewriting, HyDE, self-reflection loops to reduce hallucinations • Multi-source retrieval (databases, PDFs, Shopify, Gorgias, internal docs) 🔹 LLM Pipelines & Agentic Systems • Multi-step LLM workflows with evaluators & guardrails • LangChain, LangGraph, Semantic Kernel • Structured outputs with function calling & schema enforcement • Fine-tuned LLaMA-3 (LoRA) for domain-specific extraction & reasoning • Cost-optimized pipelines (caching, async execution, retries) 🔹 Computer Vision • HP Singapore: printer component inspection systems • Object detection, image classification, defect detection • Models deployed in real manufacturing workflows 🔹 Speech AI • Whisper fine-tuning for noisy & impaired speech • Real-time transcription pipelines • Voice cloning & synthesis (ElevenLabs-based systems) Tech Stack: Python, PyTorch, FastAPI, Docker LangChain, LangGraph, Semantic Kernel OpenAI, Claude, LLaMA-3, Whisper Azure AI Search, Azure OpenAI, Azure Document Intelligence AWS Textract, ECS, Lambda Pinecone, Neo4j, PostgreSQL Industries I’ve Worked In Banking • Healthcare • Manufacturing • E-commerce • Food-Tech Why Clients Hire Me • I’ve taken AI systems to production, not just notebooks • I understand failure modes: hallucinations, OCR noise, bad retrieval • I design systems that are scalable, debuggable, and cost-aware • I can work independently and think like an AI architect, not just a coder If you need someone who can design, build, and own an end-to-end AI system, let’s chat.

I'm an expert on machine learning with experience at major investment firms and cancer research. My experience includes, but is not limited to, time series forecasting the stock market and computer vision segmentation models that detect and pinpoint tumors in cancer patients. Technologies I know: - Python, Javascript - Pytorch, Tensorflow, SciKit-Learn, Numpy, Pandas - PostgreSQL, DynamoDB - React, FastAPI, Django, Flask - AWS EC2, AWS Lambda, AWS EKS, AWS S3

Need data extracted from complex websites, AI workflows automated, or an LLM-powered application built? I help startups and businesses automate workflows, extract data at scale, build AI-powered AI applications, and deploy reliable Python systems in production. I focus on delivering real, working systems that save time, reduce manual effort, and scale reliably. ──────────────────────────── What I Can Help You With • Saved hours of manual work through automation • Built production-ready web scraping systems • Developed AI assistants powered by GPT and RAG pipelines • Automated data pipelines processing thousands of records • Delivered reliable APIs used in production environments ──────────────────────────── AI & LLM Engineering OpenAI | Claude | Azure OpenAI | Prompt Engineering | LangChain | RAG | Vector Databases | FAISS | Pinecone | AI Agents | ChatGPT | GPT-4 | OpenAI API | Claude API | CrewAI | AutoGen | MCP (Model Context Protocol) ──────────────────────────── Python Development Python | FastAPI | Django | REST APIs | Automation | API Integration | Backend Development | Web Services Selenium | Playwright | Scrapy | BeautifulSoup | Browser Automation | Web Crawling | Data Extraction ──────────────────────────── Data Engineering ETL | Airflow | Kafka | Pandas | SQL | MongoDB | Data Pipelines | Data Processing ──────────────────────────── Cloud & DevOps AWS | Azure | Docker | CI/CD | RunPod | GPU Cloud | Deployment | Monitoring ──────────────────────────── Core Services ✔ Web Scraping & Data Extraction ✔ Browser Automation ✔ Lead Generation Systems ✔ AI Chatbots & AI Agents ✔ RAG Systems (Retrieval-Augmented Generation) ✔ GPT / LLM Integrations ✔ API Development (FastAPI) ✔ Backend Automation Systems ──────────────────────────── Why Work With Me ✔ Clean, maintainable, production-ready code ✔ Fast communication and clear updates ✔ Scalable systems built for real-world use ✔ Focus on reliability and long-term maintainability ──────────────────────────── Let’s Work Together If you're looking for a Python & AI engineer who can automate workflows, extract complex data, or build production-ready AI systems, feel free to invite me to your project.

I build production-ready AI systems that solve real business problems. With 6+ years of experience in AI/ML and Python backend engineering, I help startups and companies design, build, and deploy LLM applications, RAG pipelines, agentic workflows, NLP systems, computer vision solutions, and predictive models. My work covers Python, FastAPI, open-source LLMs, OpenAI integrations, vector search, model serving, evaluation, and cost/latency optimization. Core expertise: LLMs, RAG, Agentic AI, Python, FastAPI, NLP, Computer Vision, model deployment, API integrations, and cloud-based AI systems. If you need someone who can turn an AI idea into a reliable production system, let’s talk. Leul A

Objectives Being an enthusiastic engineer and active researcher in the field of Machine Learning and Robotics Control, I want to develop new AI-driven applications and robotic frameworks that enable humans to lead a better life style. Therefore, I am interested to become a part of an organization or a research group where I can exercise and enhance my skills that help me to achieve my research goals. I am always motivated by the challenges and try to develop something useful for the community. I believe, work until your signature becomes an autograph.

I hold an M.S. in Computer Science from Karazin National University, with a strong foundation in mathematical analysis and linear algebra, which I apply effectively to solve complex challenges in Machine Learning and Computer Vision. With 6 years of extensive commercial experience, I am an AI Engineer with a proven track record of leading development efforts and contributing to all stages of the product lifecycle, aligning technical advancements with business objectives. My expertise covers all major areas of Computer Vision, including image classification, object detection and tracking, image segmentation, and pose estimation. I have delivered impactful, end-to-end solutions in various fields, such as: - Video-driven Sports Game Action Analysis - Satellite Image Processing - AI-powered Smart Parking Systems - Real-time Video Surveillance Systems I specialize in developing custom neural network architectures for Computer Vision applications, leveraging frameworks like TensorFlow, Keras, and PyTorch, as well as tools like OpenCV, MoviePy, and FFmpeg for efficient image and video processing. My programming expertise spans Python and C++, and I have experience building CV-powered web applications using JavaScript. I have successfully contributed to projects in areas such as Sports Analytics, Parking Surveillance Systems, Satellite Image Processing, and Manufacturing & Analytics for the Beauty and Healthcare industries. Proficient in image classification, object detection, action recognition, and more, I leverage advanced neural networks to deliver effective solutions. In addition to my technical skills, I possess a strong foundation in Discrete Mathematics, Statistics, and Differential Equations, with specialized knowledge in digital signal processing, time-series analysis, and unsupervised learning. While working effectively in team environments, I also understand the importance of business logic and strategic goals. I’ve helped businesses and enterprises achieve their goals, drive financial growth, and automate processes to generate significant gains.

Want to deploy production-ready AI models, automate complex visual/text workflows, or build rigorous data pipelines? I bridge the gap between advanced research and scalable engineering. As an ML Engineer and Data Scientist with a Master’s degree, I specialize in building deterministic, high-performance AI systems. From fine-tuning LLMs and training computer vision models to engineering leakage-aware research datasets, I deliver clean, reproducible, and production-grade code. Whether you are a startup needing a custom RAG pipeline or an enterprise optimizing massive datasets, I ensure your models perform with mathematical precision. 🚀 CORE CAPABILITIES • AI Systems, LLMs & Automation Building robust RAG pipelines, structured data extraction systems, and automated workflows using Gemini, OpenAI, and Claude APIs. Expert in prompting strategies that yield deterministic, enterprise-grade outputs. • Natural Language Processing (NLP) & Dataset Engineering Experienced with Hugging Face Transformers for text classification, summarization, and multi-label categorization. Specialized in building leak-aware, balanced training/validation pipelines (e.g., CVEfixes vulnerability dataset parsing). • Computer Vision & Synthetic Data Generation Hands-on experience with object detection, image segmentation (U-Net variants), and custom preprocessing. I design synthetic image/video generation pipelines using state-of-the-art tools (Runway, Vidu, Veo, Stable Diffusion) for specialized model training and simulation datasets. • Machine Learning & Deep Learning (PyTorch & TensorFlow) End-to-end model development, validation, and optimization using PyTorch, TensorFlow, and Scikit-Learn. Expert at resolving class imbalances, handling massive datasets, and model benchmarking. • Statistical Modeling & Time Series Deep mathematical background in optimization, logistic/linear regression, hypothesis testing, and time-series forecasting. Experienced with complex real-world data analysis (e.g., NHANES health prediction models). • Data Quality & Engineering Advanced data cleaning and curation using Python, R, CleanLab, and Argilla to isolate and eliminate noise, outliers, and duplicates before they hit your training loop. 🛠️ TECHNICAL TOOLKIT • Languages: Python, R, Bash, SQL • AI/ML Frameworks: PyTorch, TensorFlow, Scikit-Learn, Hugging Face • Data Tools: CleanLab, Argilla, Pandas, NumPy, Excel (Pivot Tables) • Visualization: R (ggplot2), Python (Matplotlib, Seaborn) • Generative Media: Veo, Runway, Vidu, Stable Diffusion API Let’s turn your data into a competitive advantage. Click the "Invite to Job" button to discuss your project requirements!

About Me: I’m an AWS-certified Machine Learning Engineer with 4+ years of hands-on experience building and deploying production-grade AI and ML systems. I specialize in Large Language Model (LLM) development, Retrieval-Augmented Generation (RAG) pipelines, and Natural Language Processing (NLP) applications such as classification, summarization, and sentiment analysis. With deep expertise in AWS SageMaker and Bedrock, I help organizations move generative AI projects from proof-of-concept to production, ensuring scalability, compliance, and performance. My focus is on building reliable, maintainable AI solutions that integrate seamlessly into existing business workflows. Currently pursuing my MSc in Computer Science (AI/ML) at Georgia Tech, I combine strong theoretical foundations with proven practical skills to deliver solutions that truly drive business outcomes. Services I Offer: 1. Fine-tuning Large Language Models (LLMs) for domain-specific applications 2. Building end-to-end RAG (Retrieval-Augmented Generation) systems with vector databases (Pinecone, Bedrock KB, Opensearch ) 3. Developing AI Agents for intelligent automation and knowledge retrieval 4. Designing and deploying NLP solutions – classification, summarization, sentiment analysis, entity extraction 5. Prompt engineering and model optimization for improved performance and accuracy 6. Production ML Systems – CI/CD pipeline automation, model versioning, and continuous monitoring using AWS SageMaker and related services Skills: Programming: Python, PySpark, JavaScript, FastAPI AI/ML: LLM Development (Hugging Face, LangChain, Bedrock), NLP (spaCy, Transformers) Frameworks/Libraries: Scikit-learn, Pandas, NumPy, TensorFlow, PyTorch Cloud: AWS (SageMaker, Bedrock, Lambda, Glue, API Gateway, CloudWatch, S3) Model Lifecycle: MLflow, CI/CD for ML, Model Monitoring Databases: SQL, Pinecone (Vector DB), DynamoDB AI Optimization: Prompt Engineering, Model Fine-Tuning, Quantization Visualization: Tableau, AWS QuickSight Education: 🎓 MSc in Computer Science (AI/ML) – Georgia Institute of Technology, Atlanta, United States. 🎓 MSc in Cloud Computing (Software Engineering) – Munster Technology University, Cork, Ireland 🎓 BEng in Computer Engineering – Federal University of Technology, Owerri, Nigeria. Certifications: ✅ AWS Certified Solutions Architect – Associate ✅ AWS Certified Machine Learning – Specialty ✅ AWS Certified GenAI Developer – Professional (In progress) Industry Experience: AWS Data/ML Engineer @ HiveTekCorp (Dec 2025 - till present) MLOps Engineer @ Veterans United (Sept 2022 - Sept 2023) Data Scientist @ Veterans United (Dec 2020 - Sept 2022)

Freelancer - Verified Skills from Upwork Test System I am Muhammad Usman Ghani Khan, a seasoned Data Scientist and AI expert with over 20 𝘆𝗲𝗮𝗿𝘀 𝗼𝗳 𝗿𝗲𝘀𝗲𝗮𝗿𝗰𝗵 𝗮𝗻𝗱 𝗱𝗲𝘃𝗲𝗹𝗼𝗽𝗺𝗲𝗻𝘁 experience in machine learning, deep learning, image processing, computer vision, natural language processing, bioinformatics, medical imaging and computational linguistics. 𝗘𝘅𝗽𝗲𝗿𝘁𝗶𝘀𝗲: • 𝗔𝗿𝘁𝗶𝗳𝗶𝗰𝗶𝗮𝗹 𝗜𝗻𝘁𝗲𝗹𝗹𝗶𝗴𝗲𝗻𝗰𝗲 𝗮𝗻𝗱 𝗠𝗮𝗰𝗵𝗶𝗻𝗲 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴: I earned my PhD in Computer Science (Machine Learning/Artificial Intelligence) from The University of Sheffield, UK. With over 20 years of experience in AI, ML, and data science, I have published more than 200 impact factor publications across various AI domains, including natural language processing, computer vision, and medical applications. • 𝗖𝗼𝗺𝗽𝘂𝘁𝗲𝗿 𝗩𝗶𝘀𝗶𝗼𝗻: I am an expert in modern computer vision techniques. I have led numerous R&D projects in image processing, computer vision, and video analysis. My work includes developing machine learning models for object detection, segmentation, and image classification using 2D and 3D images. • Generative AI: I am a well versed trainer and developer working in the new paradigm of GenAI. Following are some of my expertise in the area of GenAi. Natural Language Processing including (i) text processing (ii) feature extraction (iii) word embedding and (iv) RNNs/LSTMs/GRUs Running LLM'S off the Shelf: LLM APIs, Open-source LLMs, Prompt engineering, Structuring outputs Working with vector databases like Chroma, Pinecone, Milvus etc Retrieval Augmented Generation using LangChain and LlamaIndex Prompt Engineering: Few Shot Prompting, Chain-of-thoughts Prompting Supervised finetuning via LoRA and QLoRA Quantization and Model Distillation Agentic Development: LangGraph and Crew AI etc. Deployment of LLMs: LM Studio, LangServe, AWS EC2 Instance etc. Monitoring of LLMs: LangSmith • 𝗡𝗮𝘁𝘂𝗿𝗮𝗹 𝗟𝗮𝗻𝗴𝘂𝗮𝗴𝗲 𝗣𝗿𝗼𝗰𝗲𝘀𝘀𝗶𝗻𝗴 (𝗡𝗟𝗣): I have a deep understanding of NLP techniques and large language models like ChatGPT. My experience includes developing custom NLP solutions for text generation, classification, and information extraction. I have successfully applied NLP to various domains, including biomedical text processing and computational linguistics. • 𝗠𝗲𝗱𝗶𝗰𝗮𝗹 𝗔𝗽𝗽𝗹𝗶𝗰𝗮𝘁𝗶𝗼𝗻𝘀: With extensive experience in bioinformatics and medical imaging, I develop AI solutions for healthcare, including the analysis of medical images (X-ray, CT, MRI) and biomedical data. My work spans from data preprocessing and feature engineering to model development and deployment. • 𝗗𝗲𝗲𝗽 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝗙𝗿𝗮𝗺𝗲𝘄𝗼𝗿𝗸𝘀: Proficient in TensorFlow and PyTorch, I have hands-on experience in building and training custom neural networks for various applications. My expertise includes using these frameworks for computer vision, NLP, and other machine learning tasks. • 𝗥𝗲𝘀𝗲𝗮𝗿𝗰𝗵 𝗮𝗻𝗱 𝗗𝗲𝘃𝗲𝗹𝗼𝗽𝗺𝗲𝗻𝘁 𝗟𝗲𝗮𝗱𝗲𝗿𝘀𝗵𝗶𝗽: As the Director of National Centre of Artificial Intelligence and founder of multiple research labs, including the Computer Vision & Machine Learning Lab, Bioinformatics Lab, Virtual Reality & Gaming Lab, Data Science Lab, and Software Systems Research Lab, I lead cutting-edge research and innovative projects. • 𝗧𝗲𝗰𝗵𝗻𝗶𝗰𝗮𝗹 𝗣𝗿𝗼𝗳𝗶𝗰𝗶𝗲𝗻𝗰𝘆: Experienced in Python, Java, JSON, Django, Asp.Net, R, MATLAB, and Ubuntu. Skilled in using GitHub for version control and LaTeX for documentation. 𝗘𝗱𝘂𝗰𝗮𝘁𝗶𝗼𝗻𝗮𝗹 𝗖𝗼𝗻𝘁𝗿𝗶𝗯𝘂𝘁𝗶𝗼𝗻𝘀: I have created and published freely available video lectures on YouTube for courses in Bioinformatics, Image Processing, Data Mining & Data Science, and Computer Programming. • 𝗠𝘆 𝗩𝗶𝘀𝗶𝗼𝗻: Together we grow! • 𝗠𝘆 𝗠𝗶𝘀𝘀𝗶𝗼𝗻: To be the best for your business or product!