I am an innovative AI Systems Architect with 5+ years of experience engineering end-to-end intelligence solutions. I bridge the gap between deep data science, autonomous agentic workflows, and high-performance web applications.
- Agentic AI & Generative AI: Architecting production-grade RAG-based LLM pipelines, autonomous multi-agent frameworks, and voice-driven agentic workflows.
- Full-Stack & Intelligent Systems: Building robust Next.js and React interfaces backed by complex automated data pipelines, custom APIs, and secure RBAC systems.
- Advanced Data Science: Developing diagnostic machine learning models, time-series forecasting, and high-integrity data enrichment engines.
| Category | Technologies |
|---|---|
| Languages | Python, SQL, TypeScript, JavaScript, HTML5/CSS3, DAX |
| Web Dev & Frameworks | Next.js, React.js, Tailwind CSS, Node.js |
| Generative AI & LLMs | LangChain, RAG Architecture, Hugging Face Transformers, DeepSeek, Qwen |
| Machine Learning | Scikit-learn, PyTorch, TensorFlow, Keras, XGBoost |
| Data Architecture & BI | PostgreSQL, MySQL, SQLite, Power BI, Tableau, Looker Studio |
| DevOps & Automation | AWS, Docker, Git/GitHub, CI/CD, n8n, Zapier |
- AI-Powered Multimodal Media Archive * A production-ready RAG system featuring audio transcription with timestamps, PDF text extraction, and a unified vector store with a strict citation attribution engine.
- Corrective RAG Agentic Workflow
- An advanced self-assessing AI system that evaluates the relevance of retrieved documents prior to LLM generation to eliminate hallucinations.
- MCP-Powered Voice Agent
- An intelligent, voice-driven framework leveraging the Model Context Protocol (MCP) to interactively query databases, APIs, and web sources in real-time.
- MCP-Powered Financial Analyst Agent
- A CrewAI-driven multi-agent system that parses complex financial queries, automatically generates clean Python code, and produces actionable stock visualizations.
I am highly open to collaborating on:
- Advanced Generative AI & LLM architectures (RAG, agentic workflows, MCP implementations).
- High-performance Full-Stack AI products utilizing Next.js, React, and Python microservices.
- Large-scale clinical, financial, or e-commerce predictive analytics and automated data pipelines.
- Email:isaacadewolekehinde@gmail.com
- LinkedIn:Adewole Kehinde Isaac
- Pronouns: Mr.
- ⚡ Fun fact: I design, build, and deploy the entire technical stack—from training the ML models to launching the live web platforms they run on. (And yes, I am very good at what I do!)