In my most recent project, I developed a tennis prediction engine for Oddsmaster, an AI sports betting SaaS, achieving 79% accuracy across 1,754 historical matches to support more profitable betting strategies. I'm an AI Engineer with 4+ years of experience building production-grade AI agents, RAG systems, and automation workflows for startups and businesses. I build end-to-end: from architecture and model development to API integration, cloud deployment, and multi-agent orchestration using LangGraph, LangChain, and CrewAI. WHAT I BUILD ✅ AI Agents & Multi-Agent Systems - Autonomous agents that monitor processes, use tools, manage state and memory, handle failures, and escalate to humans when needed. Built with LangGraph, LlamaIndex, OpenAI, Anthropic (Claude), and Gemini. ✅ RAG & Knowledge Systems - Retrieval-Augmented Generation pipelines with vector databases (pgvector and similar), embeddings, semantic search, and MCP (Model Context Protocol) integration for grounded, trustworthy outputs. ✅ Production AI APIs & Backend Systems - FastAPI, Docker, PostgreSQL, Supabase, and cloud deployment (AWS, Azure, GCP) with a focus on reliability, monitoring/observability, and CI/CD. ✅ Machine Learning & Computer Vision - Classification, forecasting, feature engineering, OCR, and image classification with data-leakage prevention and rigorous evaluation. ✅ LLM Fine-Tuning - LoRA, Hugging Face Transformers, and Meta Llama models for domain-specific applications. ✅ Workflow Automation & Integrations - Connecting AI systems to CRMs, ERPs, Shopify, and third-party APIs to automate real business processes, with structured outputs, function/tool calling, and human-in-the-loop guardrails. RECENT PROJECTS ✅ NFL Prediction Engine - Developed and deployed an NFL sport prediction system achieving 70% accuracy on 285+ historical games using Scikit-Learn and AWS EC2. Engineered 50+ feature pipelines with data-leakage prevention, improving predictive performance by 8%, and implemented AWS Kinesis streaming for real-time data processing. ✅ Tomato Disease Detection AI System — Deep learning computer vision model achieving 92% accuracy using AWS SageMaker, with automated CI/CD pipelines for deployment and testing. ✅ Verifi — Fine-tuned Meta-Llama-3-8B-Instruct with LoRA for financial question answering. ✅ ScaleSense — AI recruitment agent that ranks CVs against job descriptions, explains every recommendation, and helps HR teams identify top candidates in minutes. ✅ Telmu Bot — AI customer support chatbot for the Base blockchain ecosystem, deployed on Discord and Telegram using Python, LlamaIndex, FastAPI, and Docker. WHY WORK WITH ME I focus on production reliability, not just getting a model to work once, but architecting for evals, error handling, and scale. I'm NDA-compliant, communicate clearly throughout the project. If you need an AI agent, RAG system, or automation workflow built right the first time, let's talk.…In my most recent project, I developed a tennis prediction engine for Oddsmaster, an AI sports betting SaaS, achieving 79% accuracy across 1,754 historical matches to support more profitable betting strategies. I'm an AI Engineer with 4+ years of experience building production-grade AI agents, RAG systems, and automation workflows for startups and businesses. I build end-to-end: from architecture and model development to API integration, cloud deployment, and multi-agent orchestration using LangGraph, LangChain, and CrewAI. WHAT I BUILD ✅ AI Agents & Multi-Agent Systems - Autonomous agents that monitor processes, use tools, manage state and memory, handle failures, and escalate to humans when needed. Built with LangGraph, LlamaIndex, OpenAI, Anthropic (Claude), and Gemini. ✅ RAG & Knowledge Systems - Retrieval-Augmented Generation pipelines with vector databases (pgvector and similar), embeddings, semantic search, and MCP (Model Context Protocol) integration for grounded, trustworthy outputs. ✅ Production AI APIs & Backend Systems - FastAPI, Docker, PostgreSQL, Supabase, and cloud deployment (AWS, Azure, GCP) with a focus on reliability, monitoring/observability, and CI/CD. ✅ Machine Learning & Computer Vision - Classification, forecasting, feature engineering, OCR, and image classification with data-leakage prevention and rigorous evaluation. ✅ LLM Fine-Tuning - LoRA, Hugging Face Transformers, and Meta Llama models for domain-specific applications. ✅ Workflow Automation & Integrations - Connecting AI systems to CRMs, ERPs, Shopify, and third-party APIs to automate real business processes, with structured outputs, function/tool calling, and human-in-the-loop guardrails. RECENT PROJECTS ✅ NFL Prediction Engine - Developed and deployed an NFL sport prediction system achieving 70% accuracy on 285+ historical games using Scikit-Learn and AWS EC2. Engineered 50+ feature pipelines with data-leakage prevention, improving predictive performance by 8%, and implemented AWS Kinesis streaming for real-time data processing. ✅ Tomato Disease Detection AI System — Deep learning computer vision model achieving 92% accuracy using AWS SageMaker, with automated CI/CD pipelines for deployment and testing. ✅ Verifi — Fine-tuned Meta-Llama-3-8B-Instruct with LoRA for financial question answering. ✅ ScaleSense — AI recruitment agent that ranks CVs against job descriptions, explains every recommendation, and helps HR teams identify top candidates in minutes. ✅ Telmu Bot — AI customer support chatbot for the Base blockchain ecosystem, deployed on Discord and Telegram using Python, LlamaIndex, FastAPI, and Docker. WHY WORK WITH ME I focus on production reliability, not just getting a model to work once, but architecting for evals, error handling, and scale. I'm NDA-compliant, communicate clearly throughout the project. If you need an AI agent, RAG system, or automation workflow built right the first time, let's talk.WWWW…