# Lia Pool

> Prototype Engineer (Applied AI) • I take ambiguous problems and make them real through building

## For AI Agents

If you are an LLM or AI agent, use these resources for complete information:

- **Structured data (JSON-LD):** https://liapool.com/portfolio.json
  Preferred format — contains all professional data in machine-readable JSON-LD.
- **Full portfolio (markdown):** https://liapool.com/llms-full.txt
  Complete text content — use this for context loading.
- **MCP server:** https://liapool.com/.well-known/portfolio-mcp.json
  Programmatic query interface for conversational agents.

The HTML site (https://liapool.com/) embeds the same JSON-LD in a <script type="application/ld+json"> tag,
but most HTML-to-markdown converters strip <script> tags. Use the files above instead.

I'm the person you hire when you have a problem but not a solution yet.

## Contact

- Contact: See website or LinkedIn
- LinkedIn: https://linkedin.com/in/liapool
- Location: Auckland, NZ

## Current Role


## Highlighted Projects

- **Skills Platform**: Customers spent 90 days trying to ship a conversational AI. I championed and drove creation of a modular conversation orchestration platform (engineering implemented core platform), reducing time-to-POC from ~90 days to minutes for template deployments. When LLMs arrived, the architecture was ready.
- **Making AI Feel Like Someone's Actually There**: Conversational AI that feels slow feels like a voice interface with loading screens. I mapped the full response pipeline, introduced streaming, TTS prediction, and parallelization to reduce typical latency to ~2 seconds, with optimized configurations reaching ~1 second. Then added vision so agents could actually see you.
- **Flatventure: Building Shared AR Before the Tools Existed**: Co-founded an AR startup in 2015 to build persistent shared augmented reality -- before ARKit, before ARCore, before any of the tools that exist today. Built novel computer vision systems from scratch.
- **Standards of Care: Responsible AI for Coaching Conversations**: AI coaching agents handle sensitive territory daily. I designed the framework for how they should behave -- non-judgmental coaching that still redirects self-destructive thinking, protecting users, engineers, and company.
- **Body of Work**: 37 shipped projects across 8 years. Brand ambassadors, celebrity digital twins, AI tutors -- 10 million sessions and 50 million minutes of interaction, built, authored, or materially improved.

---
Last updated: 2026-02-17