Enterprise AI Architect with deep expertise in generative AI systems, cloud-native architecture, and the full ML model lifecycle. Recent accomplishments include designing RAG/GraphRAG architectures with vector databases (ChromaDB, Neo4j, FAISS), deploying multi-agent AI systems using GCP ADK with Agent-to-Agent (A2A) patterns, and architecting LLM-powered security analysis platforms with Gemini 3.0.
Skilled at defining integration patterns across LLMs, vector databases, APIs, microservices, and event-driven workflows. Technical leadership experience spans architecture governance, cross-functional design reviews, and mentoring engineering teams on AI best practices. Committed to embedding responsible AI, data governance, and security requirements into scalable, production-ready solutions.
My technical skills include:
Multi-agent AI system for energy market analysis featuring DAG-based workflow orchestration, PyTorch LSTM-based ML models with MLflow tracking, and graph analysis using DuckDB with Property Graph Query extensions. Implements MCP server for Claude Code integration enabling natural language database queries.
mlpartnership.comMedical Billing System with React frontend (Cloudfront), Python API server (Lambda), and Sqlite DB. Also Audit tab use Claude (Sonnet 4) and other LLMs to audit claim.
Medical Billing System - AWSMedical Billing System with React frontend, Python API server, and Postgres DB. Also Audit tab use Ollama LLM to audit claim.
Medical Billing System - Linux/Mac MiniRetrieval Augmented Generation systems running on Google Cloud Run.
GCP Cloud Run - RAGRetrieval Augmented Generation systems for enterprise knowledge management, combining vector databases with LLM capabilities. (Simple resume RAG and two ACTUARIAL STANDARD OF PRACTICE (ASOP) RAG with different embedding models.)
RAG ResumeContext/Cache Augmented Generation systems for enterprise knowledge management, combining vector databases with LLM capabilities. ( ACTUARIAL STANDARD OF PRACTICE (ASOP) CAG with different embedding models.)
Context Augmented GenerationMy professional resume detailing 20+ years of experience in enterprise architecture and technical leadership.
View ResumeOpen Web UI installed on Dell Zeon 64 GB RAM and Mac Mini M4 16 GB to compare response time for ollama modules.
Dell Linux ServerSample React JS demo.
React SampleMedical Billing System with React frontend, Python API server, and Postgres DB. Also Audit tab use Claude (Bedrock)) to audit claim.
Medical Billing System - AWSProject duong.casa site is an example of continuous integration and continuous delivery (CI/CD).
Github ProjectA custom-built enterprise chatbot leveraging Ollama's open-source AI models for natural language processing and contextual responses.
Explore ProjectModel Context Protocol system with (MCP) Java SDK to retrieve time.
More info on MCP...Distributed ML training setup combining high-memory batch processing with GPU-accelerated inference
Data pipelines, ETL processing, and large batch ML training
ML inference and model serving - 3-4x faster than CPU-only training
Have a question? Fill out the form below and I'll get back to you as soon as possible.
