Pat Pham / AI

Technical Product and Program Manager building local-first AI tools and custom personal agents.

Selected Works

Atlas

Live beta

A private place recommendation engine that aggregates Google Takeout history and group taste profiles into explainable suggestions.

  • Aggregates friend location signals to beat anonymous public review averages
  • Caches Google Places API data locally in SQLite to reduce lookup overhead
React Express SQLite OpenRouter +2 more

PatOnPoints

Live

An award travel search engine that routes between flight availability lookups and RAG-based booking advice.

  • Combines flight availability, cash price benchmarking, and booking instructions in one flow
  • Uses embeddings-based routing to show structured tables or RAG chat based on query intent
Next.js Express SQLite Duffel API +2 more

OpenUsage Contributions

Active contribution

Local database parsing and API wrappers for tracking usage metrics and limits across developer AI providers.

  • Provides unified visibility for developers running multiple AI coding tools
  • Parses local SQLite databases directly to show session token costs
Tauri Rust TypeScript Connect-RPC +1 more

Baymax Household Chief of Staff

In active household use

A persistent household operating layer built on Hermes Agent, combining routed models, layered memory, deterministic data pipelines, and human-approved action loops.

  • Acts as a persistent chief of staff across calendars, health, newborn routines, finances, household milestones, and developer workflows
  • Protects prompt cache by disabling advanced tools on chat gateways and isolating concurrent classification
Hermes Agent Honcho LightRAG Langfuse +4 more

Countergravity

Prototype in use

A mobile web interface and secure Tailscale tunnel for monitoring and steering local AI development environments.

  • Bridges desktop coding workspaces to mobile screens for on-the-go review
  • Secures local port connections via a private Tailscale tunnel
JavaScript Caddy Tailscale PWA +1 more

PCloud

Tinkering & Benchmarking

A personal cloud of locally hosted LLMs and automated benchmarking pipeline on dedicated remote hardware.

  • Automates remote llama-server starts and power caps on dedicated RTX 2080 Ti hardware
  • Isolates model completions and chat endpoints to prevent benchmarking parser errors
Python llama.cpp lm-eval nvidia-smi +1 more
Deep Dive →
Private Source

System Capabilities

Shared Agentic Skills

To avoid reinventing workflows, API connectors, and security configurations for every new assistant, I maintain a centralized repository of custom skills. These modular scripts and system prompts are loaded dynamically across my runtime environments.

patppham/skillz-n-thingz

Centralized Agent Skills Repository

System prompts, environment scripts, and API connectors maintained globally for local agentic workflows.

View Repository → MIT License
CAPABILITY.01

Planning & Design

Requirement specification planners, repository bootstrappers, interaction mockup systems, marketing screenshot editors, and canonical Edward Tufte visualization plugins.

plan init designer design-taste-frontend lazyweb-design app-store-screenshots tufte
CAPABILITY.02

Development & Review

Context-preserving session logging, pre-push verification checks, and multi-language code review guidelines.

historian push reviewer
CAPABILITY.03

Testing & QA

Headless browser testing pipelines, user interface visual regression audits, and background desktop OS GUI automation.

playwright-cli ui-tester computer-use
CAPABILITY.04

Infrastructure & Ops

Automated setups for cloud VPS deployments, Hostinger domain/DNS configurations, remote headless PC connections, and local LLM serving/fine-tuning.

vps hostinger local-pc local-model
CAPABILITY.05

Household Ops

Google Workspace productivity automation and personal biometrics/health API synchronization.

gog ghealth