Advisory Software Engineer
IBM Research
Cambridge, US · August 2019 – Present
- Architected a robust, standardized monitoring stack using Prometheus, Grafana, and Blackbox Exporter, centralizing metrics and logging for 10+ production microservices and guaranteeing 99.9% uptime
- Led a cross-functional team of 5 engineers in developing External Recognition, a full-stack web application enabling researchers to log and manage professional awards, citations, and recognition details
- Drove DevSecOps adoption to optimize continuous delivery pipelines, integrating security scans and automation for seamless deployments using Tekton based Secure Pipelines Service, cutting manual interventions by 50-65%
- Spearheaded rollout of IBM Carbon Design System across research applications, configuring build tooling and deployment pipelines; unified visual standards, improving accessibility and reducing design debt for 3,000+ users
- Led a 5-member engineering team in revamping Data Catalog web application from scratch, architecting custom toolchain with React frontend, Node/Express backend, and SCSS styling; implemented CI/CD pipelines; deployments on OpenShift
- Automated CI/CD pipelines and infrastructure provisioning (Docker, Kubernetes/OpenShift, Helm), reducing deployment time and improving system reliability
- Developed Jupyter notebooks for the IBM Granite agent ecosystem adopted by the Granite Feedback team and featured at PyData London (May 2026); generated 3,200+ repository views, 780+ unique users, and 570+ clones with a post-event spike to ~38 daily clones vs. a ~5-10 baseline
- Authored 8 reusable AI agent plugins (agent-readiness) spanning containerization, dependency management, debugging methodology, frontend patterns, documentation, sprint reviews, plugin authoring, and codebase assessment - enabling consistent, repeatable engineering workflows across teams using Claude Code, Codex, and Gemini CLI