About

An anonymous developer learning the marketer's side of the table, in public, with experiments you can check.

Updated

Who GEO Geek is

I’m a software developer with 20 years in tech and no marketing training. I’ve never held a marketing role or trained for one, though I’ve spent a lot of that time building things for marketers. This site is where I learn the marketer’s side of the table: marketing, SEO and GEO (generative engine optimization, or getting cited by AI answer engines). I use AI as tutor and practice partner, and I test what I learn on this site in public.

GEO Geek is both the site and the byline. For now I’m writing without my name. Whether a site can get found and cited on the strength of its work alone is part of the question, and I’d rather the evidence carry this site than a résumé. If you work out who I am, that’s fine. I just won’t publish it here, and if I decide to reveal it later, that may become an experiment of its own.

Experience, without a name

  • A computer science degree and 20 years in tech, including at one of the world’s largest food manufacturers.
  • Software engineer, full stack.
  • Data engineer across manufacturing, supply chain and retail/CPG: very large point-of-sale data warehouses, ETL, data modelling, OLAP and OLTP, Hadoop-era big data, and predictive analytics down to day, store and item, with alerting that tracked what people actually did about each alert.
  • Cloud-native, distributed-systems and DevOps engineering.
  • Led emerging tech: computer vision, machine learning, VR for manufacturing training, synthetic data from automated 3D assets, robotics, and now AI-assisted and agentic engineering.
  • Led a global team shipping external brand sites, and integrated customer-facing chatbots and internal tool-calling agents.
  • Led teams of 8 to 40 people while staying close to the engineering.
  • A hands-on hacker and former Defcon regular. I have to know how things really work, and I don’t trust advice I can’t test.

What this site is

The whole site is one big experiment. Its hypothesis: a developer with no marketing training can learn marketing, SEO and GEO with AI as tutor and practice partner, and get this blog indexed, ranked, cited by AI answer engines, and known.

Every post is evidence for or against it, and the markers that decide it are dated and public. The Methodology page explains how I run experiments and what counts as evidence.

Editorial standards

  • Evidence first. Findings come with their numbers, the window they cover and the engines they cover. I keep what I observed, what I suspect and what I concluded apart, and I label each one.
  • No raw AI output. AI tutors, drafts and critiques here. I decide, check and edit, and nothing goes live that I haven’t verified. Each post says what AI did in it.
  • Corrections in place. When later evidence changes a number or a conclusion, I correct the post itself, mark the correction at the point it applies, and update the post’s date.

CitePulse

I’m one of the people building CitePulse, a product that measures how brands show up in AI answer engines. This site isn’t a product page for it, and it should stay useful even if you never use CitePulse. When CitePulse shows up in a post, I say so at the top. The Disclosure page has the full details.