Blog

Field notes from 9 years of shipping backend systems, trimming cloud bills, and putting agents in front of real users. Specific over general.

What the AI engineering market actually wants in 2026

I read a lot of job postings and hiring reports this autumn to recalibrate my own stack. The signal is clear and a little humbling: agents, retrieval, evals and shipping to real infrastructure, with model hosting and fine-tuning further down than the hype suggests.

4 min read · AI, Career, Agents
AG-UI: the missing protocol between your agent and your frontend
3 min read · AI, Agents, AG-UI
A2UI: when the agent builds the interface
3 min read · AI, Agents, A2UI
Evals are the product: how I ship LLM features without guessing
3 min read · AI, Evals, LLMs
The favour that became a revenue line
3 min read · Automation, Python, Career
Fine-tune, retrieve, or just prompt? A decision I keep re-making
3 min read · AI, Fine-tuning, RAG
Shipping an AI copilot from a hackathon to production
4 min read · AI, LLMs, RAG
From 3.5 hours to 20 minutes: rewriting a nightly SKU sync
3 min read · Java, Performance, Concurrency
How we cut infrastructure cost by ~60% with spot instances and Kubernetes
3 min read · Kubernetes, AWS, Cost
Dropping Selenium made our crawler 15× faster
3 min read · Python, Scraping, AWS

Older posts are on Medium. Subscribe via RSS.