Mayank Malik
mayankmalik.bsky.social
Mayank Malik
@mayankmalik.bsky.social
Foodie, gardener, AI developer.
If you’re building with AI or just curious how to get better results from prompting, this is a cool toolkit.

prompts.chat

#AI #Prompting
prompts.chat
This repo includes ChatGPT prompt curation to use ChatGPT and other LLM tools better.
prompts.chat
October 1, 2025 at 12:54 AM
Just found these OpenAI Prompt Packs 👀
Basically ready-made prompts for different jobs (PMs, engineers, execs, HR, etc). Super handy if you don’t wanna start from scratch every time.

academy.openai.com/public/tags/...

#AI #OpenAI #PromptEngineering
Prompt Packs | OpenAI Academy
Unlock the new opportunities of the AI era by equipping yourself with the knowledge and skills to harness artificial intelligence effectively.
academy.openai.com
September 29, 2025 at 7:14 AM
The phrase LLMs say to me most:
“You’re absolutely right.”

Of course, I am.

Nothing like being flattered by a statistically confident yes man trained on the entire internet. #sycophantic #LLM #devlife #snark
September 19, 2025 at 3:23 AM
Vibe coding is wild. Some days I’m a 10x dev riding the high of pure flow, other days I’m just babysitting AI that hallucinates package names. 😅 #devlife #coding
September 19, 2025 at 3:11 AM
If WebSearch and WebFetch are turned on in Claude Code, it can clone any website or call any API you want it to.
Here's the result when I asked it to `create an instagram clone`. #anthropic #claude-code
September 18, 2025 at 12:32 AM
Coding with agents will quickly show you where your product thinking is half-baked. The agent will do something “wrong,” and you realize you never actually told it what “right” was. Maybe because you hadn’t fully figured it out yourself. #codingwithagents #copilot
September 18, 2025 at 12:01 AM
If you’re raising serious money for your startup, you’ll need a data room. Think of it as a neatly organized folder that gives investors easy access to everything they need.

Investors love clarity. A strong data room = less back and forth, more trust.

🔑 Steps to set one up:
September 16, 2025 at 11:44 PM
Efficiency is not a good strategy, especially in early-stage startups. Staying too small makes a startup fragile and vulnerable.
September 15, 2025 at 6:27 AM
People often ask about my AI stack. Currently, I'm using:
1. Gemini for everyday tasks and research
2. Claude for coding
3. Perplexity for search
4. Manus for creating training materials, curriculum
5. Runway for video generation
6. Gamma for slides
7. ElevenLabs for voice experiments
September 15, 2025 at 4:59 AM
If you are a recent grad aspiring to be an AI engineer, these are the 3 code labs you should showcase in your portfolio:

1. Google's Agent Stack in Action shorturl.at/hYEPN
2. Create a Chat App with Vertex AI Conversation shorturl.at/oCr3W
3. Deploy a model on GKE with Nvidia NIM shorturl.at/PwZbL
Google's Agent Stack in Action: ADK, A2A, MCP on Google Cloud  |  Google Codelabs
Master multi-agent AI development with Google’s Agent Development Kit (ADK). This hands-on workshop guides you through building, orchestrating, and connecting collaborative agents using ADK, A2A (Agen...
codelabs.developers.google.com
August 11, 2025 at 9:01 PM
Reposted by Mayank Malik
Hey fam 👋

In a new blog post, I wrote about my journey from zero to RAG engineer after 1200 hours building enterprise AI.
What actually works vs. what's just hype. Hard-learned lessons so you don't repeat my mistakes 👇
news.ycombinator.com/item?id=4474...

#TechBlog #BuildInPublic #RAG #AI #LLM
From zero to RAG engineer: 1200 hours of lessons so you don't repeat my mistakes | Hacker News
news.ycombinator.com
July 31, 2025 at 8:05 AM
Training starts here. #RTX5090 #deeplearning #workstation
August 2, 2025 at 2:53 AM
I’m a big fan of Denny Zhou. Where others use complex terms and jargon, he explains in a way that is accessible to many. youtu.be/ebnX5Ur1hBk
#LLMreasoning
youtu.be
July 29, 2025 at 1:51 AM
The authoritative guide to AI coding at work just dropped by #anthropic.

www-cdn.anthropic.com/58284b19e702...
www-cdn.anthropic.com
June 7, 2025 at 6:52 AM
Claude 4 is not a version upgrade. It’s a new category! #claude #anthropic
May 28, 2025 at 3:00 AM
This paper is a little old 😅 but its methodology is still relevant for assessing AI's labor market impact. A must-read for understanding how LLMs intersect with the workforce. arxiv.org/pdf/2303.10130
#LLMs #futureofeverything
arxiv.org
May 12, 2025 at 6:49 AM
The more people prompt, the more everyone converges to the same ideas, quality, and tone. If your AI strategy is to get good at prompting, you're regressing towards the mean. #templated-thinking #clones #AI
May 8, 2025 at 2:36 AM
Labor-replacing AI will dramatically increase capital's ability to buy results.
May 6, 2025 at 3:34 AM
canceling my subscription.
May 2, 2025 at 8:09 PM
You can now connect your React application to any MCP server and it only takes a single command! 🚀

`$ npx copilotkit@latest init -m MCP`

#copilotkit #aiengineering #buildwithAI
May 2, 2025 at 6:05 PM
When it comes to AI coding, there are two camps of engineers. 1️⃣Those who spend money to save time, and 2️⃣ those who spend time to save money. I advocate for camp 1️⃣. #AI #buildwithAI #vibecoding
April 28, 2025 at 6:12 AM
Claude expands with Max plan, Google Workspace integration, and Research feature. #AI #anthropic #claude

www.anthropic.com/news/research
Claude takes research to new places
Today, we’re introducing two new capabilities that make Claude a more informed and capable collaborator — Research and a Google Workspace integration that connects your email, calendar and documents t...
www.anthropic.com
April 26, 2025 at 11:59 PM
With `tariff`, you can finally punish your Python packages for being too convenient—slap ‘em with import tariffs and watch them crawl in like they’re going through customs.
#Python #DevHumor

github.com/hxu296/tariff
GitHub - hxu296/tariff: The official repository for tariff
The official repository for tariff. Contribute to hxu296/tariff development by creating an account on GitHub.
github.com
April 21, 2025 at 6:51 PM