Building intelligent, autonomous agents that think, plan, and act —
grounded in Python, Shell scripting, and Linux infrastructure.
- Agentic AI Systems — Multi-agent orchestration, autonomous reasoning pipelines, LLM tool-use, and AI workflow automation using frameworks like LangChain, LangGraph, and the Claude / OpenAI APIs
- Infrastructure for AI — Cloud-native platforms (AWS, Kubernetes, Docker) that agents run reliably on, including CI/CD and IaC with Terraform
- Automation with Python & Shell — End-to-end scripting from lightweight Bash glue code to full Python automation frameworks for DevOps and AI workflows
- Linux at the Core — Systems-level expertise underpinning everything: process management, networking, security hardening, and performance tuning
AI & Agents
Shell & Systems
Cloud & DevOps
