Writing about ecommerce, AI, leadership, and building at the intersection of commerce and technology.
Most ecommerce teams are past the 'should we try AI?' phase. The real question now is how to deploy it without breaking the things that already work.
AI-generated product copy is table stakes now. The real value is in how you feed the model and what you do with the output.
Autonomous AI agents are handling more retail operations than ever. But the companies seeing real results are the ones who figured out where humans still belong in the loop.
A practical look at where AI agents are delivering real results in ecommerce today, from inventory to fraud detection, and what's still hype.
Why machine learning demand forecasting works better when you pair it with human judgment, not when you try to replace it.
AI shopping agents are buying on behalf of humans. Here's what that means for how you structure product data, pricing, and the buy path.
AI agents are already discovering, comparing, and purchasing products on behalf of consumers. Here's what mid-market retailers need to know.
How generative AI is changing the retail merchandising game for real.
AI agents are transforming retail operations. Here's what's actually working and what to watch.
AI-generated product content is changing how retailers approach content. Here's what's working and what to watch out for.
I built an iOS app that gives you one clear next step when your brain won't cooperate. It's called Grounded, and it just shipped to the App Store.
Engram scored 0% on preference tracking benchmarks despite having the data. The problem wasn't storage—it was reformulation. Here's the engineering path to fixing it.
Most productivity apps help you start tasks. Done helps you finish them. Here's why I built a task closer app designed for ADHD and AuDHD brains.
OpenClaw Engram v9 adds multiple search backends, stronger recall controls, benchmarking, and local LLM support. Here's what changed and why it matters.
Choosing an OpenClaw memory plugin starts with the right criteria, not the brand name. Here is what to evaluate and why Engram fits my needs.
OpenClaw forgetting decisions and context is a memory architecture problem, not a prompt problem. Here is how to fix it with a better memory layer.
The most common OpenClaw memory complaints: agents forgetting context, compaction eating data, messy memory files, and too many fragmented setup options.
After years of meaning to stay informed about ecommerce but never having time, I built an AI-powered daily podcast that does it for me.
I built OpenClaw Engram to fix the biggest AI agent problem: forgetting everything. Here is why memory matters and what shipped in v9.
I moved joshuawarren.com from Ghost to Astro on Cloudflare Pages. Here's what drove the switch, what I gained, and what I would do differently next time.