Tech. Shit posting. Reply Girl.

Joined August 2025
Developers and testers reality
291
3,368
314
38,980
2,344,003
Moniiii retweeted
OpenAI's Decisions API reached general availability on October 22. It's designed for fast, bounded decisions inside apps and agent workflows, rather than handling every task with a general-purpose model. Where would you use it in a real backend?
4
1
7
193
Moniiii retweeted
People who started coding with AI 3 months ago are already predicting the future of software engineering.
51
5
59
1,078
OpenAI's Decisions API reached general availability on October 22. It's designed for fast, bounded decisions inside apps and agent workflows, rather than handling every task with a general-purpose model. Where would you use it in a real backend?
4
1
7
193
Moniiii retweeted
GitHub Copilot now supports server-to-server authentication for organization-installed GitHub Apps in public preview. Would you let an AI agent access company repositories without a human user being signed in? What controls would you requir
1
1
45
Moniiii retweeted
BEFORE AND AFTER CODING c++ javascript java python
8
2
18
514
GitHub Copilot now supports server-to-server authentication for organization-installed GitHub Apps in public preview. Would you let an AI agent access company repositories without a human user being signed in? What controls would you requir
1
1
45
Moniiii retweeted
be honest, can you write code without ai... 👀
27
1
26
826
BEFORE AND AFTER CODING c++ javascript java python
8
2
18
514
Moniiii retweeted
I'm already tired of tomorrow
8
2
26
458
I'm already tired of tomorrow
8
2
26
458
Moniiii retweeted
I once quit a job. During the exit interview, HR looked genuinely shocked by what I said. She’d been there for 9 years. Apparently, nobody had ever told her. Sometimes exit interviews reveal what employees were too afraid to say.
8
2
15
541
Moniiii retweeted
GPT-6.1 Terra is now available in GitHub Copilot for everyday agentic coding and terminal work, with a gradual rollout. Which real coding task would you use to test whether a new AI model is actually better?
6
2
7
169
I once quit a job. During the exit interview, HR looked genuinely shocked by what I said. She’d been there for 9 years. Apparently, nobody had ever told her. Sometimes exit interviews reveal what employees were too afraid to say.
8
2
15
541
GPT-6.1 Terra is now available in GitHub Copilot for everyday agentic coding and terminal work, with a gradual rollout. Which real coding task would you use to test whether a new AI model is actually better?
6
2
7
169
Moniiii retweeted
What actually kills startups? A. Bad product B. No distribution C. Wrong timing D. Founder ego
77
2
43
2,220
Moniiii retweeted
AI can build the feature, but the moment it breaks in production.. suddenly everyone wants a software engineer.
46
3
1
49
1,525
Moniiii retweeted
Imagine you can see one metric for every AI startup in the world. - Revenue - Retention - Inference cost - Number of paying customers - Percentage of tasks completed successfully Which metric would reveal the most about who's actually winning?
18
3
25
591
What actually kills startups? A. Bad product B. No distribution C. Wrong timing D. Founder ego
77
2
43
2,220
Moniiii retweeted
If a non-technical founder can build your entire product using AI, your biggest problem might not be AI. It might be that your product was never that complicated!
30
3
34
783
Moniiii retweeted
HTML vs CSS
26
7
3
185
2,527