DEV Community

Pragmatic AI Tools Series' Articles

Back to Debashish Ghosal's Series
Your Flaky Tests Aren't Flaky
Cover image for Your Flaky Tests Aren't Flaky

Your Flaky Tests Aren't Flaky

Comments
6 min read
I Planned 10 LLM Evaluation Experiments And Only Ran 1. It Was Enough.
Cover image for I Planned 10 LLM Evaluation Experiments And Only Ran 1. It Was Enough.

I Planned 10 LLM Evaluation Experiments And Only Ran 1. It Was Enough.

5
Comments 2
13 min read
I Thought Building Agent Observability Was a Detector Problem. I Was Wrong.
Cover image for I Thought Building Agent Observability Was a Detector Problem. I Was Wrong.

Real production data broke 35 detectors instantly

I Thought Building Agent Observability Was a Detector Problem. I Was Wrong.

27
Comments 19
8 min read
I Built Scenario Packs for Agent Regression Testing. The Integration, Not the Judge, Broke Me.
Cover image for I Built Scenario Packs for Agent Regression Testing. The Integration, Not the Judge, Broke Me.

Adapters need conformance suites too

I Built Scenario Packs for Agent Regression Testing. The Integration, Not the Judge, Broke Me.

18
Comments 21
15 min read
I Stopped Trusting AI Agents With Tools. So I Built a Gatekeeper.
Cover image for I Stopped Trusting AI Agents With Tools. So I Built a Gatekeeper.

83 real agents tested zero mocks

I Stopped Trusting AI Agents With Tools. So I Built a Gatekeeper.

55
Comments 59
14 min read
I Shipped an Agent Gatekeeper (v0.1). 14 Developers Showed Me What I Missed. Here's v0.2 — a Control Plane.
Cover image for I Shipped an Agent Gatekeeper (v0.1). 14 Developers Showed Me What I Missed. Here's v0.2 — a Control Plane.

I Shipped an Agent Gatekeeper (v0.1). 14 Developers Showed Me What I Missed. Here's v0.2 — a Control Plane.

8
Comments 2
13 min read
9 Bugs That All Looked Like a Working System
Cover image for 9 Bugs That All Looked Like a Working System

Silent statistical traps in self-improving loops

9 Bugs That All Looked Like a Working System

19
Comments 14
10 min read
I Built an AI That Rewrites Its Own Prompts — Its Safety Gate Rejected Every Single Edit
Cover image for I Built an AI That Rewrites Its Own Prompts — Its Safety Gate Rejected Every Single Edit

I Built an AI That Rewrites Its Own Prompts — Its Safety Gate Rejected Every Single Edit

24
Comments 4
8 min read
I Let an LLM Rewrite Its Own Prompt. The Real Win Was the Gate That Rejected It.
Cover image for I Let an LLM Rewrite Its Own Prompt. The Real Win Was the Gate That Rejected It.

I Let an LLM Rewrite Its Own Prompt. The Real Win Was the Gate That Rejected It.

11
Comments 1
6 min read
My Agent Found Real Improvements. The Statistics Still Killed the Promotion.
Cover image for My Agent Found Real Improvements. The Statistics Still Killed the Promotion.

My Agent Found Real Improvements. The Statistics Still Killed the Promotion.

11
Comments 2
6 min read
I Tried 4 Models to Save My Self-Improving Agent. All 4 Failed.
Cover image for I Tried 4 Models to Save My Self-Improving Agent. All 4 Failed.

I Tried 4 Models to Save My Self-Improving Agent. All 4 Failed.

24
Comments 1
7 min read