DEV Community

Mirza Iqbal profile picture

Mirza Iqbal

n8n Ambassador and Verified Expert Partner. Mastra Agent Ambassador. Clay Creator. v0 Ambassador. 17+ years in enterprise automation, €5.5M+ saved for DACH clients incl. BMW and Telekom.

Education

Harvard, Passau, IIUI

Pronouns

He/Him

Work

Automating Ops

Green tests, broken docs. Dogfooding caught what 15 tests missed

Green tests, broken docs. Dogfooding caught what 15 tests missed

Comments
3 min read

Want to connect with Mirza Iqbal?

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

Already have an account? Sign in
I keep shipping my best work to an audience of one

I keep shipping my best work to an audience of one

1
Comments
2 min read
Four ambassador titles and I still could not press publish

Four ambassador titles and I still could not press publish

Comments
3 min read
3 patterns broke when I ran Claude Code unattended for 7 days

3 patterns broke when I ran Claude Code unattended for 7 days

Comments
4 min read
Your MCP server's write tools are silently dropping their body

Your MCP server's write tools are silently dropping their body

1
Comments
3 min read
Your private GitHub repo CI is one quota away from going red

Your private GitHub repo CI is one quota away from going red

Comments
4 min read
5 ways your CLAUDE.md rules quietly fail

5 ways your CLAUDE.md rules quietly fail

1
Comments 1
4 min read
Gemma 4 makes on-device multimodal AI good enough to ship

Gemma 4 makes on-device multimodal AI good enough to ship

Comments
4 min read
Stop blaming the prompt for your agent bugs

Stop blaming the prompt for your agent bugs

1
Comments 2
4 min read
The 20 percent automation cannot close is the part that pays you

The 20 percent automation cannot close is the part that pays you

2
Comments
3 min read
Why my agent forgets the project. Memory tier mismatch.

Why my agent forgets the project. Memory tier mismatch.

Comments 2
4 min read
GPT cost failure "enterprise teams" must address immediately in week two!

GPT cost failure "enterprise teams" must address immediately in week two!

Comments 1
3 min read
The agent setup that updates itself. What self-evolving actually means in practice

The agent setup that updates itself. What self-evolving actually means in practice

Comments
4 min read
AI reported success.

AI reported success.

Comments
4 min read
A weekly tool-use test is the only signal

A weekly tool-use test is the only signal

Comments
3 min read
The 5 hidden ways your Claude Code bill quietly doubles

The 5 hidden ways your Claude Code bill quietly doubles

Comments 2
4 min read
Opus 4.8 barely moved the leaderboard. It moved the one number that decides if your agents can be trusted.

Opus 4.8 barely moved the leaderboard. It moved the one number that decides if your agents can be trusted.

Comments
5 min read
Claude Code's workflow docs are a menu.

Claude Code's workflow docs are a menu.

Comments 2
4 min read
16 days until Anthropic's Agent SDK billing splits.

16 days until Anthropic's Agent SDK billing splits.

Comments
4 min read
Why your agent agrees with everything. The pushback gap.

Why your agent agrees with everything. The pushback gap.

Comments
4 min read
Agent memory poisoning. The 4-stage enterprise damage chain.

Agent memory poisoning. The 4-stage enterprise damage chain.

Comments 8
5 min read
The pattern that keeps long-running jobs alive through cap exhaustion

The pattern that keeps long-running jobs alive through cap exhaustion

Comments
3 min read
I have 160 Claude Code hooks. The 5 that pay rent every day.

I have 160 Claude Code hooks. The 5 that pay rent every day.

Comments
4 min read
5 ways AI agents quietly die inside n8n production

5 ways AI agents quietly die inside n8n production

Comments
4 min read
€40 n8n vs 28% weekly Anthropic quota. Which /goal layer should you actually run?

€40 n8n vs 28% weekly Anthropic quota. Which /goal layer should you actually run?

Comments
6 min read
loading...