HNHacker News
TopNewBestAskShowJobs

bhagyeshsp

15 karma · joined June 15, 2022

Self-taught developer, hacker, tech-enthusiast and business owner.

What am I up to: https://bhagyeshpathak.com/now

submissionscomments
bhagyeshsp··on Using LLM for confidence scoring is a bad idea
The author has used weather prediction confidence as an example. I think it is not analogous to LLM's next-token prediction. LLM's next-token is predicted from a finite statistical model. And for the weather prediction model, there is no finite data-based statistical model.

I was surprised that the author hasn't mentioned Jev-like models. They're quite good for general purpose confidence rating.

bhagyeshsp··on Ask HN: Who is using MCP in production?
I'm using in production for one of my physical products. The product is a set of 50 printed cards for facilitation. The MCP server can be used by the facilitator to plan their workshop. The reason behind providing MCP is simple: my end-users do not need traditional API, they need something that they can install in their Claude Desktop app and work with the proprietary data of the product. My product's data lives at edge in Cloudflare Workers. (Of course, a determined user can scrap the data if they want, but that's fine.)

Other uses of other people's production MCP I'm using is Dodo Payments' MCP server. It has helped me a ton for looking up their documentation etc. 10xed my OpenCode agent's speed in building payment integration. This was the first serious usage and benefit I experienced of MCP servers.

Reference: My production MCP server: https://www.npmjs.com/package/@sisyphus-consulting/wuwei-mcp

bhagyeshsp··on Show HN: I scratched my own itch by solving a niche Basecamp problem
In case it is not clear, try the demo at https://app.feedhammer.com/demo
bhagyeshsp··on How LEDs are made (2014)
I guess I had read this article 12-13 years back. I think it was this same article.

One of the things I vaguely remember was reading somewhere that working on this LED manufacturing severely damages the workers' eyes. I don't know how much of it is true and if it is, whether that is still the case.

bhagyeshsp··on Child marriages plunged when girls stayed in school in Nigeria
Looks like I mistook an ideologue zealot for a rational being.

Bbbye

bhagyeshsp··on How are you handling prompt injection across multi-step agent workflows?
This is interesting. I want to understand how much of a latency does Omega Walls introduce to the workflow? Or is it recommended for long-running workflows where a couple of additional seconds of latency is fine?

I've built a simple AI-chat interface for one of my products and I had experienced a latency of 6000 ms by adding a router LLM to selectively inject context to the main LLM. Dropped that idea completely.

EDIT: my bad, just read the term "multi-agent workflow" in the title. Still, some idea about the added latency would be helpful.

bhagyeshsp··on Child marriages plunged when girls stayed in school in Nigeria
Let us take your previous comment as the basis

> Birth rates would also improve when boys and men are educated.

There is no evidence of this being true. This is certainly a narrative peddled by many ideologues.

> Both genders need education and child support programs.

Poorest of poor and illiterate people happen to have more children than the rest.

> Men/Boys need to understand what responsibilities they have, if they choose to have a child.

If men are educated on responsibilities of alimony and child support, with almost no rights, they would neither marry nor have legitimate children.

> They also need to understand the effects that having a child has on a woman's body.

This maybe your personal dream and that's fine. But this has nothing to do with the topic at hand.

> Governments around the world would benefit their society by investing in family planning, family support (esp. child care) to enable parents to work and provide for their family.

Family-planning is euphemism for reducing children per woman. There's no benefit of having less children -> leading to less economic activity in the future. The family support you keep touting about is moot point. Government does not have their own money. People pay taxes which are used by government.

> An educated and healthy populace (from infant to old age) benefits everyone.

Agreed on this point. The definition of benefits are subjective but overall, it is agreed that it is a net positive.

bhagyeshsp··on Child marriages plunged when girls stayed in school in Nigeria
This is almost the opposite of what happens.

The more educated/developed a nation, the lesser their birth rate is going to be.

I understand the "shoulds" but that's not what the data suggests.

In essence, we can't have the pie and at the same time eat it.

The most useful thing education does for children is reduce child-mortality rate.[1]

Sources: https://raphael-godefroy.github.io/pdfs/mali_final.pdf

[1] https://eric.ed.gov/?id=ED503923

bhagyeshsp··on Vibe coding and agentic engineering are getting closer than I'd like
"YOU'RE A SENIOR SOFTWARE ENGINEER!!!"
bhagyeshsp··on Vibe coding and agentic engineering are getting closer than I'd like
> The thing that really helps me is thinking back to when I’ve worked at larger organizations where I’ve been an engineering manager. Other teams are building software that my team depends on.

> If another team hands over something and says, “hey, this is the image resize service, here’s how to use it to resize your images”... I’m not going to go and read every line of code that they wrote.

The distance of accountability of the output from its producer is an important metric. Who will be held accountable for which output: that's important to maintain and not feel the "guilt".

So, organizations would need to focus on better and more granular building incentives and punishment mechanisms for large-scale software projects.

bhagyeshsp··on Show HN: Curl.md – URL to Markdown for Agents
This is quite interesting. I'm going to try it.

If I understand correctly, my agent just needs to run `curl curl.md/targetsite.com` and it will receive the webpage in an optimized Markdown file format.

Is there any specific depth until which curl.md will crawl?

bhagyeshsp··on Texico: Learn the principles of programming without even touching a computer
Though it looks like it is made for kids, adults can learn a lott from these. Superb resource. Thanks
bhagyeshsp··on Agentic Coding Is a Trap
Okay, after re-reading the thread, looks like, this question was not in bad faith. While commenting, I was reading from bottom-to-top and so made an opinion about you based on your bottom thread comments (which still stands correct), but for other readers, they deserve an answer, not the questioner.

Anyone wondering about my proficiency, I can code without internet or AI help. But it takes enormous amount of time and mistakes.

bhagyeshsp··on Agentic Coding Is a Trap
Dude, pls stop. You're making me laugh. Seriously.

This is my last response to you. Expended your credits.

bhagyeshsp··on Agentic Coding Is a Trap
Thanks for the heads-up. Appreciate that.
bhagyeshsp··on Agentic Coding Is a Trap
My native language is Gujarati. Done my schooling and college in Gujarati too.

Absolutely, I understand what you're saying.

One of the things people miss out, in most of the discussions is that they think "if you were really serious, you would have figured it out". I agree with that in most instances but language and skill acquisition is a complex process as everyone knows.

English being the de-facto reservoir of programming knowledge and applications, it takes substantial amount of time and effort to cross the threshold of understanding and transference.

In any case, I'm an eternal optimist and I believe in action. It was a great experience listening to people's opinion here and I was kind of shocked to find that some of them are so siloed in their chambers, that's interesting nonetheless.

bhagyeshsp··on Agentic Coding Is a Trap
Repeating what I said on the previous comment:

---

Now you want to have a "discourse"? No thanks.

Honestly, I want you to live with your assumptions and beliefs. No more goodwill from my side to your username.

bhagyeshsp··on Agentic Coding Is a Trap
Now you want to have a "discourse"? No thanks.

Honestly, I want you to live with your assumptions and beliefs. No more goodwill from my side to your username.

bhagyeshsp··on Agentic Coding Is a Trap
Attention Please Everyone

You're looking at a specimen of one of the most rattled species nowadays. It is so fun finding them under these articles.

Their last attempts at finding inner-validation.

What a waste.

bhagyeshsp··on Agentic Coding Is a Trap
Haha..so many assumptions. And all of them being objectively wrong.

Cope harder. Try harder to demoralize. It is not going to work.

I had heard HackerNews had some of this loser bunch of personalities. I didn't expect them to show themselves so soon.

But hey, I realized "AI bashing" articles are the best places to find these gate-keepers. Makes sense now.

bhagyeshsp··on Agentic Coding Is a Trap
I don't believe in victimhood and so didn't want to go here, but since we are comparing notes:

English alphabets came into my education at the age of 10. I got my first computer at the age of 21. I began speaking broken English around the age of 23. Proper internet at the age of 25 or so.

Not to mention, my native language doesn't have programming books, even today.

Of course, an avid reader and Science nerd here. Curiosity and tinkering never stopped.

bhagyeshsp··on Agentic Coding Is a Trap
Pls get rid of your assumptions, then we can talk.
bhagyeshsp··on Agentic Coding Is a Trap
The answer is similar to: whose code is easier to untangle: human-A or human-B?

"Depends".

bhagyeshsp··on Agentic Coding Is a Trap
Learning the fundamentals of programming and their translation to code. I'm decent at engineering, managing and marketing solutions.
bhagyeshsp··on Agentic Coding Is a Trap
Self-taught, "junior" here.

Due to English-language limitation my most adult life, I struggled to code. Used visual coding etc. But of course, I can't make a living on drag-and-drop harness.

Comes in GPT-3.5, accelerated my learning. Now I'm running my incorporated company, just launched one software-hardware hybrid product. Second one is a micro-SaaS in closed beta.

The point is: when people use "juniors" as a fixed shaped blobs of matter, they focus on the juniors that were in any case going to make mistakes: AI or not. Misses the key point of agentic usage.

bhagyeshsp··on Technical Debt of AI Systems: Agent Runtime
I was unaware of these many layers and different number of primitives. Great read, provided some great ideas.

I'm trying to build a small, hobby agentic system locally, looks like it is an order of magnitude more difficult to create your own full secure agent than I had previously thought.

bhagyeshsp··on We stopped hiring engineers for coding ability
I have noticed that such hyped up announcements-cum-validation-seeking articles don't actually answer anything specific.

For example, they can't tell you in one sentence what they replaced the "coding ability" screening with.

More enshitification.

I understand the AI-aided productivity gains but don't take people for fools. Running AI models in a sandbox environment takes technical know-how. Companies never hired people just for coding ability.

bhagyeshsp··on Uber torches 2026 AI budget on Claude Code in four months
Hahaha.. good one :D
bhagyeshsp··on Uber torches 2026 AI budget on Claude Code in four months
I didn't know this. There's a term for this--which everyone of us now know--enshitification.
bhagyeshsp··on Uber torches 2026 AI budget on Claude Code in four months
Wonderful, so when will I see novel features in my Uber app?
Page 1 of 2Next →