You are reading the simple version of this page. Switch to standard view
Data & AI engineering · San Diego

I help computers keep information neat so they can do smart things that don't break.

Pine Marten is John Houghton's own company. John has helped big companies like Meta, Apple, Kaiser Permanente and Intuit with their information for ten years. He builds helpers that sort information, count it, and think with it — and some smart helpers that stay inside your building so your secrets stay safe.

Engraving-style pine marten illustration
Martes martes
Services

How I can help

Data engineering

I build pipes that carry information to one big tidy box, check that it's right, and show your team how to take care of it.

dbtDagsterDuckDBSQL

AI engineering

I build chatty computer helpers that do real jobs, test them to make sure they're good, and give them safety rules from the start.

LLMsagentsevals

Local & private AI

Smart computer helpers that can listen, see and talk, living on your own computer so your information never goes outside.

on-premSTT/TTSvision

Analytics & decision support

I count things, try tests, and make easy charts so the bosses get answers to their real questions.

metricsexperimentation

Asynchronous quantum cryptography systems

Your secrets are both safe and not safe at the same time, until you look. I will get back to you about which. Eventually.

post-quantumpre-revenueidempotent decoherence
How we work

Small team, lots of practice, I do the work

Assess

First I look at your information and what you want, then I write down a plan of what to do first.

Build

I build with your tools, a little bit at a time, and check and explain each piece.

Hand off

I leave instructions and teach your team, so they can keep building without me.

John Houghton
About

Hi, I'm John.

My job is taking messy information and making it neat and trustworthy. Most recently I helped teams at Meta called FAIR and Reality Labs, and before that Apple Maps, Kaiser Permanente and Intuit. I studied economics, so I care about the choices people make with the information.

I build computer systems for information and AI in San Diego. I studied economics and I love to build things. You can hire me to help, or for a full-time job.

Portfolio

Things I made

SD Open Data Pipeline

It takes San Diego's free city information, tidies it up, and figures out which neighborhood area each thing belongs to — and checks itself every time.

DagsterdbtDuckDB spatialCI
sd-open-data-pipeline

LLM Dataset Curation

It cleans up the reading material for AI: removes copies, throws out junk and mean words, hides private details, and makes a report card.

LLMsdata quality
llm-dataset-curation

Vision Lab · Local AI Showcase

A computer that watches through cameras and talks with you, all inside one Mac — it remembers what happens, checks its work, and hides private things.

on-devicevisionvoice
vision-lab

Finance Wizard

A money helper app where a computer helper does the work, with fun charts to try "what if" ideas.

FastAPIReactagents
finance-wizard-demo

Personal Finance Lake

A home computer box that collects bank info, labels each purchase, and checks that everything adds up every night.

DuckDBdbtlaunchd
finance-lake-demo

Multi-Agent Assistant

A team of computer helpers that look things up, plan your day, take notes and check for safety — and they always ask a person before sending anything.

agentsorchestration
multi-agent-assistant

Distributed Multimedia ETL Pipeline

It sorts through lots and lots of pictures using many computers at once, throws out copies and look-alikes, makes the files smaller, and keeps them neatly organized.

SparkDelta LakeAirflow
Distributed-Multimedia-ETL-Pipeline

Multi-Provider LangChain

A website where you can pick which AI brain to talk to, by typing, drawing, or talking out loud.

LangChainFastAPIvoice
multi-provider-langchain
Experience

Places I've worked

Meta (contract) — Senior Data Engineer · 2022–2026Built and ran the pipelines that delivered multi-terabyte image, video and multimodal training datasets to AI researchers at Meta Superintelligence Labs (FAIR), owning quality, rights filtering and self-serve tooling. Also owned the executive metric layer for Reality Labs’ VR strategy, the source of truth for Quest and Horizon usage and revenue. Assignment concluded September 2026.
Apple Maps (contract) — Senior Analytics Data Engineer · 2021–2022Led the move of Maps’ production reporting from Hive to Snowflake with zero disruption to downstream teams, and owned a self-serve analytics layer of 50+ pipelines and 25 dashboards.
Kaiser Permanente (contract) — Data Engineer · 2018–2021Built the executive metrics platform for a regulated healthcare organization, with automated anomaly detection so issues were caught in the pipeline, not in the boardroom.
Intuit — Senior Manager, Marketing Analytics · 2017Directed a team of analysts and engineers delivering a unified 360° customer view across nine data sources, setting the data-quality standards it ran on.
Background

Education & training

B.A. Economics

San Diego State University, 2011

Machine Learning Specialization

University of Washington, 2017

Data Mining Certificate

UC San Diego Extension, 2017

Interests

Outside the terminal

Local-first and private AI, economics and markets, endurance sports, and rare books.

Do you have an information puzzle?

Write me a little note about what you're doing. I'll write back by the next work day.