20+ years building, designing, and operating mission-critical infrastructure, platforms, and distributed systems. Currently at DigitalOcean engineering the systems that scale the cloud. I blend platform architecture with site reliability, treating operational engineering as a software discipline—building custom developer tooling, fail-safe automation frameworks, and intelligent CLI utilities that eliminate system toil, optimize architecture, and give engineering teams back their time.
Projects
A modular CLI built for DigitalOcean's CloudOps team to safely manage bare-metal server operations — power control, hypervisor admin, spare-capacity discovery, VM/event diagnostics, alert-response automation, and full droplet lifecycle management up to permanent destruction — from a single interface, replacing fragmented dashboards, ephemeral automation containers, and one-off scripts.
- Safe path = fast path: read-only diagnostics run freely, state changes on production hardware require peer-review confirmation, and irreversible actions (permanently destroying droplets) require a second, distinct typed confirmation on top of that
- Executes safety-verified power workflows directly over a REST API instead of an ephemeral container session — with automatic, interactively-gated fallback and retry when a safety check blocks the run
- Crash-recoverable power-cycle detects interrupted runs and restores hardware rather than leaving it half-configured
- Spare-capacity finder chains inventory search with live placement verification, flagging unverifiable results rather than guessing
- Credentials pulled from a password manager on demand, never persisted; guided menu mode surfaces the equivalent direct command after every action, teaching the CLI instead of hiding it
Bash
CLI/Systems Design
GraphQL/REST
SSH Automation
Infrastructure Safety Engineering
Cross-Platform
Architected a persistent VS Code workspace utility consolidating local shell environments, AWS profiles, and markdown playbooks into a single sidebar dashboard. Engineered a fail-safe terminal stream layer that pastes scripts without automatic execution to eliminate human-error risks during critical on-call outages.
VS Code Extension API
TypeScript
Shell
AWS
Markdown
AI-powered job discovery and resume tailoring pipeline. Fetches jobs from Adzuna and direct job boards, scores them against your resume using local TF-IDF and domain overlap scoring, and generates ATS-tailored resumes via Claude API. Self-hostable with a FastAPI dashboard, Docker packaging, user accounts, and run quota management.
Python
FastAPI
SQLite
SQLAlchemy
Claude API
Adzuna API
Docker
JWT
Started as a Kubernetes resource optimiser. Evolved into a comprehensive infrastructure intelligence platform that maps, analyses, and optimises cost structures across multi-cloud and on-premise VMs, storage, and networking. Serves DevOps, CloudOps, FinOps, Billing, and CTO/CFO teams with automated savings snapshots — without compromising SRE-grade operational safety.
Python
Kubernetes
Multi-cloud
FinOps
Cost optimisation
A high-performance graph computation engine built for the KOIP platform, designed to process complex knowledge graph traversals at scale. Part of KOIP
DAG
Graph theory
Infrastructure topology
Cross-platform IDE-to-browser toolchain that bridges local developer workspaces with web-based LLM chat interfaces. Syncs state and execution hooks between a local VS Code extension and a custom Chrome extension.
Node.js
TypeScript
VS Code Extension API
Chrome Extension
Automates Operational Availability Review compliance across 1,200+ microservices at DigitalOcean — replacing a manual, error-prone process with a continuously running platform.
- Ingests a live service catalog hourly, validates documentation links, and checks for required operability metadata (PagerDuty, Jira, ownership)
- Per-service scoring with monthly health snapshots to track compliance trends over time
- Read-only Streamlit dashboard with filtering, pagination, and multi-format exports
Python
PostgreSQL
SQLAlchemy
Streamlit
Docker
Kubernetes
Vault
Consolidates 8–12 parallel SSH terminals and copy-paste workflows into a single CLI and lightweight web dashboard for hypervisor incident response at DigitalOcean.
- YAML-driven playbooks with automated escalation: SSH → serial console → IPMI → BMC → PDU
- Real-time integration with Atlantis, StackStorm, Slack, and PagerDuty
Python
FastAPI
Click
asyncio
Docker
VPN-deployed