DEV Community

Debashish Ghosal profile picture

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/

Location Seattle, WA Joined Joined on  Personal website https://github.com/deghosal-2026/ github website

Pronouns

he/him

Work

Engineering Leadership

Top 7
3
Two Year Club
8 Week Community Wellness Streak
4 Week Community Wellness Streak
2 Week Community Wellness Streak
1 Week Community Wellness Streak
Writing Debut
One Year Club
A Test Run I Aborted Came Back and Corrupted My Evidence

A Test Run I Aborted Came Back and Corrupted My Evidence

11
Comments 2
5 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.

Already have an account? Sign in
Our Agent Gate Blocked Every Bad Change. So Everyone Started Rubber-Stamping It.

Our Agent Gate Blocked Every Bad Change. So Everyone Started Rubber-Stamping It.

15
Comments 7
3 min read
Your Agent's Config Is a Dependency With No Lockfile

Your Agent's Config Is a Dependency With No Lockfile

15
Comments 19
3 min read
The Shadow Backlog: How Agentic Code Generation Overwhelms System Architecture

The Shadow Backlog: How Agentic Code Generation Overwhelms System Architecture

21
Comments 2
8 min read
Why Your AI Agent Is Brilliant for 20 Minutes and Then Starts Lying

Roots in lost context and self-correction failure

Why Your AI Agent Is Brilliant for 20 Minutes and Then Starts Lying

16
Comments 12
4 min read
We Moved to a Bigger Model to Debug the Test. Then We Went Back to the Smaller One.

We Moved to a Bigger Model to Debug the Test. Then We Went Back to the Smaller One.

14
Comments 2
5 min read
Nobody Writes About Retiring an Agent. Here's the Checklist.

Nobody Writes About Retiring an Agent. Here's the Checklist.

12
Comments 5
4 min read
What Publishing 35+ Agent Articles Taught Me About Writing (and About Agents)

What Publishing 35+ Agent Articles Taught Me About Writing (and About Agents)

17
Comments 3
4 min read
Rate Limits Are a Queueing Problem: Capacity Math for Agent Fleets

Rate Limits Are a Queueing Problem: Capacity Math for Agent Fleets

15
Comments 8
4 min read
I Keep a Log of Every Rule My AI Agent Broke. The Patterns Are Uncomfortable.

Readers share similar logs of autonomous behavior

I Keep a Log of Every Rule My AI Agent Broke. The Patterns Are Uncomfortable.

23
Comments 27
5 min read
The "Magic Pill" Delusion: Why AI Licenses Won't Fix Broken Platform Engineering

The "Magic Pill" Delusion: Why AI Licenses Won't Fix Broken Platform Engineering

16
Comments 2
6 min read
MCP Went Stateless This Year. Your 'Supports MCP' Checkbox Is Now a Lie.

MCP Went Stateless This Year. Your 'Supports MCP' Checkbox Is Now a Lie.

11
Comments 2
3 min read
The FDE Mindset Shift: From Code Craftsmanship to Organizational Rewiring

The FDE Mindset Shift: From Code Craftsmanship to Organizational Rewiring

19
Comments
5 min read
97 of 99 Sounds Great. It Would Have Been a Bad Release Number.

97 of 99 Sounds Great. It Would Have Been a Bad Release Number.

11
Comments
5 min read
Debugging Is the Skill AI Made Most Expensive to Skip

Debugging Is the Skill AI Made Most Expensive to Skip

12
Comments 3
3 min read
My Agent Said 'All Tests Pass.' CI Disagreed Three Minutes Later. Both Were Honest.

My Agent Said 'All Tests Pass.' CI Disagreed Three Minutes Later. Both Were Honest.

17
Comments 5
3 min read
The Myth of the 10x Solo AI Developer: A Reality Check from Teams Actually Shipping

The Myth of the 10x Solo AI Developer: A Reality Check from Teams Actually Shipping

17
Comments 17
8 min read
I Changed One Model Pin and Ten Test Results Went Stale

I Changed One Model Pin and Ten Test Results Went Stale

10
Comments 2
5 min read
Supervising AI All Day Is Exhausting. Nobody Scheduled Relief.

Supervising AI All Day Is Exhausting. Nobody Scheduled Relief.

14
Comments 4
3 min read
My 'Second Backend' Existed Only in the README

My 'Second Backend' Existed Only in the README

14
Comments 5
4 min read
Architecting Deterministic Safety Gates for Agentic Workflows

Architecting Deterministic Safety Gates for Agentic Workflows

18
Comments 9
7 min read
10 Prompts That Turn ChatGPT Into a Senior Data Analyst

10 Prompts That Turn ChatGPT Into a Senior Data Analyst

20
Comments 1
5 min read
Claude Looked Blocked. My Policy Had Never Fired.

Claude Looked Blocked. My Policy Had Never Fired.

10
Comments 3
5 min read
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.

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.

16
Comments 4
4 min read
Stop Controlling, Start Leading: The "Let Them" Mindset at Work

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

14
Comments 3
3 min read
Open Source Maintainers Are the First People Whose Inbox Got Automated

Open Source Maintainers Are the First People Whose Inbox Got Automated

14
Comments 9
3 min read
The Soak That Reported 5,003 Deliveries for 5,000 Records

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

9
Comments 7
5 min read
Why Forward Deployed Engineers Are Suddenly Hot: The Death of the Centralized AI Guild

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

10
Comments 1
6 min read
You Don't Have a 2.0 Platform Proposal. You Have a Preference.

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

10
Comments
12 min read
Vibe Coding Is a Spike, Not a User Story

Compares AI tools to disposable XP spikes

Vibe Coding Is a Spike, Not a User Story

11
Comments 10
4 min read
My Daemon Dropped 1,000 Records and Reported Success

My Daemon Dropped 1,000 Records and Reported Success

9
Comments 11
6 min read
The Boundary Problem: When Staff Engineers Don't Draw Lines—and the Chaos That Follows

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

8
Comments
3 min read
Zero LLM Calls From a Working Agent. My Capture Hook Was Lying.

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

10
Comments
5 min read
Slopsquatting: 1 in 5 AI Code Samples Recommends a Package That Doesn't Exist

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

12
Comments 8
5 min read
92 Green Field Tests Meant Nothing Until I Counted Which Could Lie

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

9
Comments 2
5 min read
Context Engineering over Prompt Engineering: Architecting Codebases for AI Agents

Context Engineering over Prompt Engineering: Architecting Codebases for AI Agents

9
Comments 1
6 min read
The 10-Line AI Working Agreement Your Team Never Wrote

The 10-Line AI Working Agreement Your Team Never Wrote

9
Comments 19
3 min read
I OOM'd on 100 Rows Because One File Was 590 MB

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

8
Comments 4
6 min read
Three Times My Test Failed and the LLM Was Right

Exposes flaws in human testing assumptions

Three Times My Test Failed and the LLM Was Right

25
Comments 34
6 min read
Rushed vs. Rigorous: The Executive Blueprint for AI Adoption That Doesn't Alienate Your Best Engineers

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

9
Comments 11
8 min read
98 Calls, 98 Errors, 0 Detections: The Bug My Test Suite Called Green

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

14
Comments 9
6 min read
From Pilot Purgatory to Agentic Chaos: Why Medium Enterprises Keep Breaking Their Own SDLCs

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

11
Comments 3
8 min read
My CI Was Red for 255 Commits and Nobody Looked

My CI Was Red for 255 Commits and Nobody Looked

15
Comments 4
5 min read
226 Scenarios, 43 Detectors, 1 Clean Run: Building a Test Harness That Can't Lie

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

18
Comments
6 min read
The Verification Trap: Why Generative AI Is Making Senior Engineers Slower (and How to Fix It)

The cognitive tax of reading synthetic code

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

17
Comments 22
6 min read
My Helm Chart Linted Green and the Pod Crash-Looped

My Helm Chart Linted Green and the Pod Crash-Looped

11
Comments 6
5 min read
A Drift Detector That Cries Wolf Is Worse Than No Detector

A Drift Detector That Cries Wolf Is Worse Than No Detector

13
Comments 8
4 min read
The Bug Lives at the Seam, Not in the Logic

Real-world examples from a HivePlane field test

The Bug Lives at the Seam, Not in the Logic

23
Comments 16
4 min read
No Bootstrap, No Principal: The Day My Audit Log Became Fiction

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

12
Comments 10
4 min read
The 75% Code Trap: How Senior Engineers Are Actually Surviving AI Mandates

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

8
Comments 5
9 min read
I Signed a Regression and Almost Shipped It

I Signed a Regression and Almost Shipped It

9
Comments 4
4 min read
One `rm -f` Silently Disabled My Crash Detection

Marker files create hidden failure modes

One `rm -f` Silently Disabled My Crash Detection

25
Comments 11
6 min read
The Last Generation That Learned the Hard Way

The Last Generation That Learned the Hard Way

9
Comments
6 min read
My Release Pipeline Refused to Ship My Release. It Was Right.

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

7
Comments 8
5 min read
Your AI Dashboard Is Green. Your Delivery Isn't.

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

18
Comments 9
13 min read
Same AI, Four Different Jobs: Startup, Mid-Size, Big Tech, IT Services

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

7
Comments 8
7 min read
The AI Memo Went Out. Then What?

The AI Memo Went Out. Then What?

14
Comments 6
7 min read
Your CEO Sees 5x. Your Engineers See a Longer Review Queue.

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

6
Comments 1
6 min read
You Can't Test Money Controls With a Free Model

You Can't Test Money Controls With a Free Model

9
Comments 8
4 min read
Five Things Release Day Caught That Six Weeks of Green Tests Didn't

Why default packaging leaks junk files

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

17
Comments 12
5 min read
loading...