Pluggable multi-agent decision system that helps users navigate complex choices in healthcare, business, and legal domains through cognitive-emotional debate and adaptive reasoning.

A self-learning Multi-Agent CLI tool, built with the Strands Agents SDK, uses OpenAI, yfinance, and Tavily to handle complex financial queries and features a self-correcting ticker lookup loop.

Advanced agentic patterns lab: Multi-agent orchestration with Claude Agent SDK and agent-to-agent communication with Strands SDK

Technical implementation of high-reasoning AI agents using the AWS Strands SDK, featuring autonomous tool-use, Model Context Protocol (MCP) integration, and multi-agent coordination.

Enterprise-grade GenAI platform for financial services. Dual-profile security (customer vs employee), 4-layer Bedrock Guardrails enforcement, hybrid RAG from PDF policies + Notion FAQs, and brand-consistent NLG synthesis. Built on AWS Bedrock + Claude 3.7 Sonnet.

A modular, code-first AI customer support agent for Auto Insurance. Built using the Strands SDK and Claude 3.5 Sonnet to demonstrate end-to-end intent validation, policy generation, and secure checkout flows using Python and Flask.

An autonomous Agent-to-Agent (A2A) orchestration hub built with Strands SDK, Amazon Bedrock, and TypeScript. Features self-healing triage, episodic memory, and an LLM-as-Judge evaluation pipeline.