New on The Quality Forge: "The Portable Orchestra"
forge-quality.dev/articles/por...
01.03.2026 18:40
👍 0
🔁 0
💬 0
📌 0
New on The Quality Forge: "The Portable Orchestra"
forge-quality.dev/articles/por...
Databases that lost their data. Features that were never real. And a conductor who was drowning grief in work and didn't notice until the frustration came out in ALL CAPS toward an LLM.
Venting at an agent doesn't improve its output. It degrades your instructions.
forge-quality.dev/articles/con...
Found this today, shared by a friend, and it resonated with me. What do you think? - resonantcomputing.org
Yegge's "AI Vampire" resonated hard. AI drains your energy — agreed. But it also drains your judgment, and that's the bite nobody sees coming.
Wrote about what this looks like through a quality engineering lens on The Quality Forge: forge-quality.dev/articles/qua...
Every test passed. The system was fundamentally broken.
10 days of debugging what the tests said was working. 8 releases. 10+ forensic investigations.
The uncomfortable lesson: evidence chains beat opinions.
New on The Quality Forge 👇
forge-quality.dev/articles/cas...
What happens when you apply DDD to agentic development?
Agents go deep instead of wide. Integration is still hard. But review cycles dropped from 3-6 to 2.
14 days, 12 domains, honest lessons.
forge-quality.dev/articles/14-...
Watched Claude Code run rm -f *.db while I was reading a book.
Two months of agent learning data. Gone in a keystroke.
Except I had a backup. The project didn't.
New article: What happens when personal habit is your only safety net.
forge-quality.dev/articles/orc...
Karpathy calls it a "magnitude 9 earthquake." Beck is coding solo with a "genie." Cherny shipped 259 PRs in 30 days—all AI-written.
I spent 2025 building agentic testing platforms and learning.
New article on the mindset shift for 2026.
forge-quality.dev/articles/new...
Playing this morning with Nano banana, when I asked for 8k resolution, photorealistic textures, and dramatic lighting, I got some 🤯 results.
Last week: 5 releases in 5 days.
This week: 8 more releases.
The agents reported "100+ Q-values recorded."
The database said zero.
The orchestra took a bow. The score was empty.
New on The Quality Forge: "When the Orchestra Says 'Done' But Plays Off-Score"
🔗 forge-quality.dev/articles/orc...
Test as Transformation - AI, Risk, and the Business of Reality by Keith Klain qualityremarks.com/test-as-tran...
From AI skeptic to conducting agent swarms in one year.
New on The Quality Forge: my journey through prompt → context → agentic engineering.
The twist? Human judgment matters MORE now, not less.
Includes video from my Aveiro lecture.
🔗 forge-quality.dev/articles/tes...
Building EmotiStream in 48 Hours: An Agentic Development Journey - www.linkedin.com/pulse/buildi...
10 years ago, I walked into Agile Testing Days for the first time.
This year, I stepped onto its stage.
Volunteer. Speaker. Exhibitor. Three hats, one incredible week.
Here's my story from ATD 2025. 👇
www.linkedin.com/pulse/back-f...
Good morning from Potsdam with a sprinkle of snow. I did some counting, and this is my 10th #AgileTD as a volunteer(since 2016), and the first as a speaker. Achievement unlocked, I will be joining 🦄 on stage. Whoohoo 🌈🎉
Wednesday 26. November 2025, 16 h - Room E1, at @agiletdzone.bsky.social, join my talk, where I will share some of the lessons I learned while building two (three) open-source Agentic testing platforms using Agents.
"Transform to Agentic QE or die!"
Nah. Wrong framing entirely.
Real question: Which parts of your QE work actually benefit from agents vs traditional approaches?
New article: practical guidance from experience, not vendor hype.
Context over dogma. Always.
forge-quality.dev/articles/age...
Hard lesson learned: Knowing best practices ≠ following them.
Especially when you're stretched thin and tell yourself, "The tests are passing, probably fine."
Just wrote about shipping 5 releases with broken features, because I forgot to be a Tester:
forge-quality.dev/articles/lea...
Used AI agents to build AI agents.
Then, I used my old TypeScript QE agents to verify my new Python QE agents.
Then a human expert caught what both generations missed.
The 5-day journey shared:
forge-quality.dev/articles/orc...
AI agents claimed my Q-learning system was "fully functional."
The database had 0 rows.
The problem: optional constructor parameter = silent failure
The fix: persistence adapter pattern
The lesson: verify agent claims with actual queries
forge-quality.dev/articles/sho...
After a year of building AI-assisted and agentic testing platforms, I'm sharing where AI testing tools fail and what actually works.
forge-quality.dev/articles/ai-...
Achievement unlocked!!! I will be joining 🦄 on stage 😍
Happy to share that I will be giving a talk at Agile Testing Days, Nov. 24 - 27, 2025, sharing my experience and learnings from building two open-source testing platforms with Agents.
Hope to see you there?
agiletestingdays.com/2025/session...
Agents tried to commit code without my review. Said features were complete when they weren't. Reported using AgentDB when it was just a wrapper.
I still managed to ship in 3 days, which would typically take 6-8 weeks.
The real Agentic QE Engineering story -
www.linkedin.com/pulse/leadin...
My latest article on turning your expertise into AI-Powered Expertise using Claude Skills
www.linkedin.com/pulse/buildi...
How I Used Claude Flow Agents to Upgrade My Agentic QE Fleet (In Under 12 Hours)
www.linkedin.com/pulse/how-i-...
Yes, I am coming, looking forward seeing you soon in Potsdam at ATD
I loved listening to your interview; thanks for sharing your wisdom.
I started writing, but forgot to share here, so here it goes, my latest article: forge-quality.dev/articles/hol...
Back from Agile Testing Days 2024: A Reflection on Learning, Growth, and Community
www.linkedin.com/pulse/back-f...
What a week, a lot of learning and fun with friends, but it is time to go back home, thank you all the wonderful people at #AgileTD and hope to see you next year 🤗