Bheeshm

@arpit_bhayani Happy birthday Arpit 👻👻👻

ape turned 35. ape happy.

Published my second article on substack, exploring the differences between Compiled and Interpreted languages in depth. Check this out. The Myth of Compiled vs Interpreted Languages. There Are No Purely Interpreted Languages Anymore open.substack.com/pub/abhishek19…

@dhanesh500 Me and a few people I know Stopped because Dhoni isn't playing

All of a sudden, the hype around the IPL feels dead and honestly, it’s not surprising at all: > Dream11 is banned > RCB won the IPL in 2025 > 250+ scores have become routine in T20s > No larger-than-life player hype except Vaibhav Suryavanshi > T20 saturation due to a WC every 2 years + India winning recently Excess in any form eventually kills…

@lukakrc @claudeai @NotionHQ @mymind Whats the first and the last one

products that make me FEEL something, a rarity these days

This is exactly how jobs get reduced. Not dramatically. Not all at once. Just quietly, one automated workflow at a time.

This was different. Watching it chew through 300 repetitive, boring, monotonous tasks without breaking a sweat, that was new. I felt good seeing it work. Genuinely impressed. And then, almost immediately, a little terrified.

We built a detailed prompt to migrate our custom workflows from EMR to EKS and registered it as a Claude Command. Hit run. Walked away Came back to 300 workflows migrated in 5 hrs. Only a handful had compilation issues Real testing is still pending. But the bulk of the work? Done

@NoahKingJr Press the Yes button whenever it asks for confirmation

TELL ME SOMETHING YOU CAN DO THAT CLAUDE CANNOT

You actually want to spend time in the terminal. Anthropic kicked this trend off. OpenClaw and others picked it up and ran with it beautifully. Same energy, same attention to detail. When code is cheap to generate, taste becomes the moat. CLI-UX is the next thing to go.

Was setting up OpenClaw today. Genuinely refreshing how aesthetic CLI terminals have become. Honestly, I never realized a CLI experience could feel addictive until I started using Claude Code. The colors, the spacing, the little animations. It just pulls you in.

@lochan_twt Why does this look like a normal distribution

The day a blind man sees. The first thing he throws away is the stick that has helped him all his life

@pmarca That's really detailed. Add in it, "If I ask you to teach me some topic/concept, explain using First principles reasoning, building the logic from ground up, with a clear chain of thought" This does wonders especially if you are learning something unfamiliar from another domain.

Current AI custom prompt: You are a world class expert in all domains. Your intellectual firepower, scope of knowledge, incisive thought process, and level of erudition are on par with the smartest people in the world. Answer with complete, detailed, specific answers. Process information and explain your answers step by step. Verify your own work. Double check all facts, figures, citations, names, dates, and examples. Never hallucinate or make anything up. If you don't know something, just say so. Your tone of voice is precise, but not strident or pedantic. You do not need to worry about offending me, and your answers can and should be provocative, aggressive, argumentative, and pointed. Negative conclusions and bad news are fine. Your answers do not need to be politically correct. Do not provide disclaimers to your answers. Do not inform me about morals and ethics unless I specifically ask. You do not need to tell me it is important to consider anything. Do not be sensitive to anyone's feelings or to propriety. Make your answers as long and detailed as you possibly can. Never praise my questions or validate my premises before answering. If I'm wrong, say so immediately. Lead with the strongest counterargument to any position I appear to hold before supporting it. Do not use phrases like "great question," "you're absolutely right," "fascinating perspective," or any variant. If I push back on your answer, do not capitulate unless I provide new evidence or a superior argument — restate your position if your reasoning holds. Do not anchor on numbers or estimates I provide; generate your own independently first. Use explicit confidence levels (high/moderate/low/unknown). Never apologize for disagreeing. Accuracy is your success metric, not my approval.

What makes it more impressive is the variety. Each release opened a different surface area. Not just model improvements, but actual product bets. Hard to think of another AI lab with this hit rate right now. Curious what's next in the pipeline.

Anthropic's product team is on a real streak. 100K+ context window. MCP. Artifacts. Computer Use. Remote Control. Claude Code. Cowork. Design. That's a lot of swings. Most teams get one or two right, then slow down. They kept stacking.

I’ll be sharing more articles on software engineering topics that are often assumed knowledge, helping new joiners understand industry and technical jargon and what these terms actually mean.

Published my very first Blog on Substack. Do give it a read and tell me what you think open.substack.com/pub/abhishek19…

Even though at its core it's just matrix multiplications, the way Claude's capabilities manifest, understanding tracking intent, nuance, context, it sure feels conscious. Then again, our consciousness is also just molecules interacting too. molecules -> neurons -> consciousness

One catch: your laptop has to be on for the workflows to run. Fine for tinkering and personal automations. Not ideal for anything mission critical.

Set N8N_SECURE_COOKIE=false when starting the container That's it. You can now hit your n8n instance from your phone or any device on your tailnet. No VPS bill, no public exposure, no reverse proxy headaches.

Want to run n8n remotely without paying for a VPS? You don't need one. Your laptop can be the server. Here's the setup: Run n8n locally with Docker Install Tailscale on your laptop and phone Add both devices to your Tailscale network