Debashish Ghosal

TLDR - Engineer -> Manager -> Tech Leadership, passionate about AI, Developer experience, planet scale infra, transformation - AI-native leadership. Rest here - https://www.linkedin.com/in/deghosal/

Seattle, WA Joined on Oct 15, 2024 https://github.com/deghosal-2026/

Pronouns

he/him

Work

Engineering Leadership

Pinned

83 real agents tested zero mocks

Debashish Ghosal

Aug 13

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

#ai #agents #security #gatekeeper

14 min read

Real-world agents break clean evals

Debashish Ghosal

Aug 2

I Built an Agent Eval Harness. Real Agents Broke the Clean Version of the Story

#ai #agentskills #learning #opensource

9 min read

Debashish Ghosal

Nov 1

AI Now Writes 75% of Google's New Code. 46% of Developers Don't Trust the Output. Nobody Wants to Talk About the Second Number.

#discuss #ai #career #productivity

4 min read

Want to connect with Debashish Ghosal?

Create an account to connect with Debashish Ghosal. You can also sign in below to proceed if you already have an account.

Create Account

Already have an account? Sign in

Debashish Ghosal

Nov 1

Stop Controlling, Start Leading: The "Let Them" Mindset at Work

#healthydebate #leadership #discuss #growth

3 min read

Debashish Ghosal

Oct 30

Open Source Maintainers Are the First People Whose Inbox Got Automated

#discuss #ai #opensource #programming

3 min read

Debashish Ghosal

Oct 29

The Soak That Reported 5,003 Deliveries for 5,000 Records

#ai #testing #programming

5 min read

Debashish Ghosal

Oct 29

Why Forward Deployed Engineers Are Suddenly Hot: The Death of the Centralized AI Guild

#discuss #ai #platform #leadership

6 min read

Debashish Ghosal

Oct 29

You Don't Have a 2.0 Platform Proposal. You Have a Preference.

#discuss #devops #career #productivity

12 min read

Compares AI tools to disposable XP spikes

Debashish Ghosal

Oct 29

Vibe Coding Is a Spike, Not a User Story

#discuss #ai #webdev

4 min read

Debashish Ghosal

Oct 29

My Daemon Dropped 1,000 Records and Reported Success

#ai #testing #programming #llm

6 min read

Debashish Ghosal

Oct 28

The Boundary Problem: When Staff Engineers Don't Draw Lines—and the Chaos That Follows

#discuss #leadership #career #technicallead

3 min read

Debashish Ghosal

Oct 28

Zero LLM Calls From a Working Agent. My Capture Hook Was Lying.

#discuss #ai #testing #python

5 min read

Debashish Ghosal

Oct 27

Slopsquatting: 1 in 5 AI Code Samples Recommends a Package That Doesn't Exist

#discuss #ai #security #programming

5 min read

Debashish Ghosal

Oct 26

92 Green Field Tests Meant Nothing Until I Counted Which Could Lie

#ai #testing #programming

5 min read

Debashish Ghosal

Oct 26

Context Engineering over Prompt Engineering: Architecting Codebases for AI Agents

#discuss #ai #aiagents #softwareengineering

6 min read

Debashish Ghosal

Oct 25

The 10-Line AI Working Agreement Your Team Never Wrote

#discuss #ai #productivity #beginners

3 min read

Debashish Ghosal

Oct 25

I OOM'd on 100 Rows Because One File Was 590 MB

#ai #outofmemory #testing #llm

6 min read

Exposes flaws in human testing assumptions

Debashish Ghosal

Oct 25

Three Times My Test Failed and the LLM Was Right

#ai #programming #llm #testing

6 min read

Debashish Ghosal

Oct 24

Rushed vs. Rigorous: The Executive Blueprint for AI Adoption That Doesn't Alienate Your Best Engineers

#discuss #ai #career #productivity

8 min read

Debashish Ghosal

Oct 23

98 Calls, 98 Errors, 0 Detections: The Bug My Test Suite Called Green

#healthydebate #ai #testing #programming

6 min read

Debashish Ghosal

Oct 21

From Pilot Purgatory to Agentic Chaos: Why Medium Enterprises Keep Breaking Their Own SDLCs

#discuss #ai #devops #softwareengineering

8 min read

Debashish Ghosal

Oct 21

My CI Was Red for 255 Commits and Nobody Looked

#ai #failingci #testing

5 min read

Debashish Ghosal

Oct 19

226 Scenarios, 43 Detectors, 1 Clean Run: Building a Test Harness That Can't Lie

#healthydebate #ai #programming #testing

6 min read

The cognitive tax of reading synthetic code

Debashish Ghosal

Oct 19

The Verification Trap: Why Generative AI Is Making Senior Engineers Slower (and How to Fix It)

#discuss #ai #softwareengineering #productivity

6 min read

Debashish Ghosal

Oct 19

My Helm Chart Linted Green and the Pod Crash-Looped

#ai #testing #k8 #crash

5 min read

Debashish Ghosal

Oct 18

A Drift Detector That Cries Wolf Is Worse Than No Detector

#healthydebate #ai #testing #llm

4 min read

Real-world examples from a HivePlane field test

Debashish Ghosal

Oct 16

The Bug Lives at the Seam, Not in the Logic

#ai #testing #llm #programming

4 min read

Debashish Ghosal

Oct 15

No Bootstrap, No Principal: The Day My Audit Log Became Fiction

#healthydebate #ai #testing #controlplane

4 min read

Debashish Ghosal

Oct 14

The 75% Code Trap: How Senior Engineers Are Actually Surviving AI Mandates

#discuss #ai #career #productivity

9 min read

Debashish Ghosal

Oct 13

I Signed a Regression and Almost Shipped It

#ai #testing #release #llm

4 min read

Marker files create hidden failure modes

Debashish Ghosal

Oct 12

One `rm -f` Silently Disabled My Crash Detection

#healthydebate #ai #programming #testing

6 min read

Debashish Ghosal

Oct 11

The Last Generation That Learned the Hard Way

#healthydebate #ai #employeeexperience #career

6 min read

Debashish Ghosal

Oct 10

My Release Pipeline Refused to Ship My Release. It Was Right.

#ai #release #testing #programming

5 min read

Debashish Ghosal

Oct 10

Your AI Dashboard Is Green. Your Delivery Isn't.

#ai #roi #llm #agents

13 min read

Debashish Ghosal

Oct 10

Same AI, Four Different Jobs: Startup, Mid-Size, Big Tech, IT Services

#healthydebate #ai #employeeexperience #career

7 min read

Debashish Ghosal

Oct 9

The AI Memo Went Out. Then What?

#healthydebate #ai #employeeexperience #career

7 min read

Debashish Ghosal

Oct 8

Your CEO Sees 5x. Your Engineers See a Longer Review Queue.

#healthydebate #ai #employeeexperience #career

6 min read

Debashish Ghosal

Oct 6

You Can't Test Money Controls With a Free Model

#healthydebate #ai #llm #testing

4 min read

Why default packaging leaks junk files

Debashish Ghosal

Oct 6

Five Things Release Day Caught That Six Weeks of Green Tests Didn't

#ai #testing #agents #llm

5 min read

Debashish Ghosal

Oct 4

The 15-Line Test That Catches the #1 Killer of Operator Trust

#healthydebate #ai #testing #llm

6 min read

Debashish Ghosal

Oct 4

Your Transformation Isn't Failing. Your Evidence Is.

#ai #leadership #cto #employeeexperience

9 min read

Debashish Ghosal

Oct 4

I Shipped a Green Test That Lied About My Pipeline

#ai #testing #llm #productivity

4 min read

Debashish Ghosal

Oct 2

They Learned to Code Before Copilot. They're Not Anti-AI. They're Pro-Evidence.

#ai #employeeexperience #career #growth

7 min read

Debashish Ghosal

Oct 2

I Said Isolation Was Structural. Then Tenancy Shipped and Proved Me Right the Hard Way

#healthydebate #ai #multitenancy #testing

4 min read

Debashish Ghosal

Oct 2

My Model-Swap Attack Worked. The Gate Was Right — My Test Was Wrong.

#ai #modelswap #testing #llm

4 min read

Debashish Ghosal

Oct 2

Coverage Stopped Being a Code Metric. It's a Career Metric Now.

#healthydebate #learning #career #growth

7 min read

Debashish Ghosal

Oct 1

I Tried to Sneak Four Bad Agents Past My Own Certification Gate. All Four Got Blocked.

#ai #llm #testing #agents

5 min read

Debashish Ghosal

Sep 30

I Ran My Career Like a Production System for 12 Weeks. The Dashboard Was Wrong Before I Was.

#healthydebate #growth #career #leadership

7 min read

Debashish Ghosal

Sep 27

A Certification That Changes Every Run Is a Coin Flip With a Signature

#ai #testing #llm #certification

5 min read

Debashish Ghosal

Sep 27

I Built Two Agent Systems. Each One Proved the Other One Wrong.

#healthydebate #ai #llm #programming

3 min read

Debashish Ghosal

Sep 27

Stop Prompting, Start Onboarding: Treat Your AI Agent Like an Intern

#ai #learning #programming #growth

2 min read

Debashish Ghosal

Sep 26

One Hung API Call Used to Kill My 1,000-Run Benchmark. Here's the Fix.

#ai #llm #programming #benchmark

3 min read

Debashish Ghosal

Sep 26

AI Promoted Every Developer to Reviewer. Nobody Measured Whether We Got Worse.

#ai #programming #testing #codereview

5 min read

Debashish Ghosal

Sep 26

Kubernetes for Agents: Why Agent Fleets Need a Control Plane

#ai #k8 #kubernetes #control

5 min read

Debashish Ghosal

Sep 25

I Trusted My Agent Demos for Years. Then I Built a Gate That Says No.

#ai #llm #controlplane #agents

5 min read

Debashish Ghosal

Sep 24

7 Agent Eval Mistakes That Cost Me Weeks (And the One-Line Fixes That Ended Them)

#ai #evaluation #llm #programming

5 min read

Debashish Ghosal

Sep 22

I Cut 2,490 Agent Test Runs to 206 and Kept the Same Coverage

#ai #programming #testing #llm

5 min read

Debashish Ghosal

Sep 19

1,558 Tests Green and No Auth: The Tests That Never Actually Ran

#ai #security #programming #testing

3 min read

Debashish Ghosal

Sep 19

What Do You Do While AI Codes? I Make Mine Argue With Itself.

#healthydebate #ai #testing #llm

5 min read

Debashish Ghosal

Sep 18

The Bottleneck Moved From Writing Code to Proving It

#ai #llm #programming #testing

4 min read

Debashish Ghosal

Sep 17

I Let AI Plan 170 Changes. It Made the Same 3 Mistakes Every Time.

#ai #agents #llm #testing

8 min read

Comments discuss losing cognitive mental models

Debashish Ghosal

Sep 15

AI Wrote Half My Codebase. The Maintenance Bill Showed Up in Month Three.

#ai #agents #llm #maintenance

5 min read