At Trellix Insights, I build and harden cybersecurity products using React, Node.js, Python, FastAPI, AWS, and GCP. I've added AI-powered report generation with Claude Sonnet and Amazon Nova Pro through AWS Bedrock while optimizing prompts to reduce inference costs.
I've strengthened production security through CSP and Cache-Control headers, keyless IAM-based signing, secrets management, JWT token administration, and SQL injection prevention. I also built an end-to-end HashInfo revocation audit feature across React, FastAPI, and DynamoDB.
I improve reliability through unit testing across frontend applications, AWS Lambdas, REST APIs, and batch jobs using Jest, React Testing Library, pytest, Mocha, Chai, and Sinon.js. I’ve also investigated production issues and refactored components to eliminate false error notifications.
Beyond my work at Trellix, I built an AI-orchestrated remediation system using RAG, LangChain, Pinecone, FastAPI, React, Power Automate, and GitHub Actions. My IEEE-published research spans machine learning, cryptography, and explainable AI.
