AI Engineering Newsletter
AI engineering notes that earn their bandwidth
Practical AI engineering for senior developers and B2B teams. Token economics, agent architecture, eval design, and the patterns that decide whether an AI feature ships or stalls.
Roughly twice a month. No hype, no fluff, no recycled news.
Three things, every issue
Each edition focuses on one or more of these. No filler, no link roundups, no generic AI commentary.
Token economics that hold up at scale
Real numbers from production agents. RTK, prompt caching, model routing, and the cost-per-task math that decides whether an AI feature ships.
Agent architecture for B2B builds
Planner-executor-reviewer patterns, MCP integration, memory schemas, and the evals that prevent silent regressions when a model upgrades.
Senior engineering applied to AI
Twelve years of production engineering meets agentic coding. Claude Code, Cursor, and the senior practices that separate demos from systems.
A sample of what subscribers read
Newsletter editions usually expand on these long-form pieces with extra implementation notes and source code.
Stop Burning Claude Tokens: How RTK Cuts AI Coding Costs 60-90%
Why engineers blow through Claude Max in four days, and the shell-level fix.
Read the post'AI Doesn't Code' Is a Skills Problem, Not an AI Problem
Ten patterns that produce most AI frustration, each with a one-line fix.
Read the postHow I Build AI Agent Teams That Actually Ship for B2B Companies
The 8-week engagement structure for production agent systems.
Read the postClaude Max and the High-Volume Engineer
Workflow benchmarks from 12 senior engineers running Claude Max full-time.
Read the postPooya Golchian, AI engineer in Dubai
Twelve years of production software engineering, the last three focused on AI agent systems for B2B companies. I write what I am actually building, with the numbers that matter to engineering leaders making real budget decisions.
About the newsletter
What lands in your inbox, how often, and how to leave.
Practical AI engineering for senior developers and B2B technology teams. Topics span token economics, agent architecture, eval design, MCP integration, memory systems, model routing, and Next.js patterns for AI-powered products. Every issue is grounded in production work, not speculation.
Roughly twice a month, occasionally weekly when a major model release or framework change warrants faster coverage. You will never get more than four emails in a single month, and most months it will be two.
Senior engineers, engineering leaders, founders, and product owners building or buying AI features in B2B contexts. The writing assumes engineering literacy and skips foundational explanations. If you are a beginner looking for tutorials, the public blog is a better starting point.
No. The list is hosted on infrastructure I control. Your email is used only to deliver the newsletter and is never shared, sold, or used for retargeting. You can unsubscribe with one click from any issue, and unsubscribing removes the address completely.
Yes. Most newsletter editions are based on long-form blog posts published at pooyagolchian.com/blog/. Recent representative examples are linked on this page, including pieces on Claude token economics, AI agent architecture, and B2B engineering practices.
The newsletter includes condensed playbooks, behind-the-scenes engineering notes, and links to client-only material that does not appear on the public blog. Subscribers also see new posts roughly two days before they are indexed publicly.
Pooya Golchian, a senior AI engineer based in Dubai with twelve years of production software experience. He builds AI agent teams for B2B companies and writes about the engineering practices that make those builds ship reliably.