AI and Automation Trends in 2026
Practical insights into the latest AI and automation trends, including AI trading platforms, compliance management software, and workflow automation tools.
Writing / Blog
Practical write-ups on automation, LLMs, and the infrastructure that makes them production-ready.
Practical insights into the latest AI and automation trends, including AI trading platforms, compliance management software, and workflow automation tools.
The latest news and updates in AI, automation, and workflow tools, including new releases and funding announcements.
I've tried Zapier, Make, and raw Python scripts. n8n sits at a sweet spot between visual flexibility and code-level control. Here's my reasoning.
Running a local language model is easy. Running one reliably under load, with a clean API, proper auth, and logging, is a different problem.
httpx + BeautifulSoup handles 70% of jobs. For the rest — JS-rendered SPAs, bot detection, infinite scroll — Playwright async is the cleanest tool I've found.
The vector search is the easy part. Getting chunking right is where most RAG implementations quietly fail.
A walkthrough of the Gmail automation system I built — triage, priority labelling, templated replies, and the edge cases that nearly broke it.
Running batch ML workloads on EKS taught me things the documentation won't. Node pools, spot interruptions, GPU scheduling — the hard lessons.
Reliable AI agents require more engineering than prompt engineering. Tool schemas, grounding, confidence thresholds, and rollback logic — what the tutorials skip.
The patterns that separate pipelines that work from pipelines that work reliably — idempotency, observability, graceful degradation, and schema evolution.
Not everything should be automated. A framework for identifying high-ROI automation targets before writing a single line of code.