suriya-ganesh

I might.

It's not just this specific issue, honestly. Throwing wrench on all economies, that my wife and I bet on is what's horrible. Research fund cuts on premium institutes, the wonky arrests etc.

Even yesterday, I had to make a case for why all of this certainty might be worth it. And it was not easy. At this point though, I certainly agree that the US is not in a trajectory for appreciating external contributions.

Interesting decision. I'm on the F1 -> H1B pipeline myself as a software engineer. And my wife is a researcher working on Genetic Engineering.

Of the both of us, I've been the strong proponent for moving the US. and with each passing day, its getting harder to make a strong case for the pain, and uncertainty of moving here.

Lately everything has been counter to what one would expect from a pro-growth, accelerationist country. But I understand where the reasoning is coming from, though.

Maybe.

The workers are involuntary but willing participants, in a grand scheme where the queens and males get to create new generations. But this is possible only if we anthropomorphise a lot.

because at the level of ants/bees I'm not even sure what "individual" even means.

But genetically they originate from the same individual, live for the betterment of the whole, and have very minor say in what happens to themselves or their genes. Much similar to cells in a human being does.

I agree but in these cases the genetic variance that is being accumulated, is only limited to a few individuals who more or less don't interact with the world, except through their workers. But the workers themselves have no way of accumulating or passing on the genetic variation.

In some sense the genetic feedback loop for ant population is designed in such a way that, it makes sense when looking at each ant colony as a singular organism

Unlike the giraffe, or elephants, who are individually capable of accumulating genetic variations.

This is wild. But eusocial insect have a lot of bizarre eccentricities in sex determinism. less than 1% of the colony can actually reproduce, every other being is there for the betterment of the 1%. The workers will mutilate, sacrifice and kill themselves just for the queen to have 0.1% better survivability.

It is helpful to think of the whole colony as a singular organism as opposed to individuals, because our understanding of individual starts breaking down at these levels

Even with that, Live demos are incredibly more better than hour long demos.

So far, I've tried this in, chrome, firefox, safari, arc and comet. loads in all of them

Being a big server proponent myself. Usually for one reason or the other there is need to introduce some socket style communication to the frontend and that becomes impossible in a single machine after a certain threshold.

Is there something obvious that I'm missing?

Hey!! Thanks for sharing this.

Your message and the child comment really has given me a perspective that I didn't have before.

Something Very practically stoic about it.

I vibe with this entirely.

Rust to me makes a lot more sense. The compiler gives reasonable errors, the code structure is clean and it's obvious where everything should go.

I just can't deal with Typescript at all. There is a sense of uncertainty in TypeScript that is just unbearable. It is really hard to be confident about the code.

I can confirm this is exactly how llms are working. Spent two hours trying to get an llm to implement a filescan skip a specific directory. Tried, claude code, Gemini and cursor. All agents debugged and wrote code that just doesn't make sense.

Llms are really good at template tasks, writing tests, boilerplate etc. But, Most times I'm not doing implement this button. I'm doing there's a logic mismatch in my expectation

I do, but logs are better.

This tracks with how I've been doing my debug logs.

I ask the model to create tons of logs for a specific function and make sure there's an emoji in the beginning to make it unique (I know HN hates emojis).

Best thing of all, I can just say delete all logs with emojis and my patch is ready. Magical usage of LLMs

Everything can be modeled as selling shovels to the miners.

We're all building tools for other people. As long the users like a product, I think it's moot to call it shovel selling.

This is a problem that occurs across game development to ML datasets.

We built oxen to solve this problem https://github.com/Oxen-AI/Oxen (I work at Oxen.ai)

Source control for large data. Currently our biggest repository is 17 TB. would love for you to try it out. It's open source, so you can self host as well.

  Location: San Francisco CA, New York
  Remote: Yes
  Willing to relocate: Yes
  Technologies: Rust, Golang, K8s, postgres, Docker, Typescript
  Résumé/CV: https://suriya.cc/surya-resume.pdf
  Email: suriyaganesh097@gmail.com

I have 4 years of experience working on a Distributed Privacy Database (>100K TPS) , RBAC and PBAC governance systems, and Federated Learning on mobile devices (100M+ devices) . More recently I graduated from CMU in Privacy Engineering, and have been involved with finetuning LLMs for privacy use cases.

I love working on early stage products where there is a lot of ownership and iterations. Do ping me if you want someone to nerd out about how snapshot isolation is not strictly serializable.

What am I getting wrong? you planted a coconut but grew a palm tree ?

I think what the parent comment is saying is that, lobster was likely introduced as an elite/rare dish to people in the current century increasing the appeal

This tracks with my experience. We built a complex processing pipeline for an NLP classification, search and comprehension task. Using vector database of Proprietary data etc.

We ran a benchmark of our system against an LLM call and the LLM performed much better for so much cheaper, in terms of dev time, complexity, and compute. Incredible time to be in working in the space seeing traditional problems eaten away by new paradigms

I tried as well. Seems like it is a proprietary model.

I'm in the Chrome -> Arc .character arc (heh!) myself.

And, profiles is the only thing preventing me from going all in on Chrome.

I have 6 profiles across my personal, university, and work contexts. And can't imagine living without them.

likely because hassabis was a child prodigy. Was a chess master at 13. Lead Cambridge chess team. its not surmise to assume that demis had impeccable school record

In some sense this is what, Rabbit R1 tried to do. they just shipped a low end custom android skin in a new formfactor so that they own the platform. Didn't go well for them

I can answer to this, having worked on an assistant that is always on, from your phone.

The platforms (ios, Android, etc.) are very limiting. It is hard to have something always on and listening. Especially apple is aggressive with apps running in the background.

You need constant permissioning and special privileges. The exposed APIs themselves are not enough to build deep and stable integrations to the level of Siri/Google Assistant.

This is not true.

Honestly, diagnosing for ADHD accelerate the improvement.

It is impossible to fight if you don't know what you're fighting. It enables you to prevent repeated patterns, not chasing your tail in an endless struggle.

This saved me a ton when back in college in rural India without Internet in 2015. I would download whole websites from a nearby library and read at home.

I've read py4e, ostep, Pgs essays using this.

I am who I am because of httrack. Thank you

The author also has a rust based qr library that looks equally awesome

https://github.com/zhengkyl/fuqr

apropos to this.

You can pretty much google up and download millions of Aadhar directly from the web.

https://x.com/deedydas/status/1838595739137773570

I've been using marblism to build a couple of projects over the last 2 weeks. It's an incredible product.

Here are my reviews:

- The tech stack selection is exceptional. It comes with batteries included, backend, auth, permissions etc.

- I love the ability to view the database, auth, in one place.

- The LLM itself gets stuck once the APIs and their interactions become slightly complex and deviates from the happy path. (likely because even the starter code is closer 10K lines of code across different files and frameworks)

- I wonder if instead being very opinionated, just a cloud LLM environment with a slightly less "fat" template would be optimal. because after all, people who would be twiddling with generated code are going to be developers who can (most likely) decide what they want.

I've been working on browser agent the last week[1]. So this is very exciting. There are also browser agent implementations like Skyvern[2] (Also YC backed) ,or Tarsier[3] Seems like, finic is providing a way to scale/schedule these agents? If that's the case what's the advantage over something like airflow or windmill ?

If I remember correctly, Skyvern also has an implementation of scaling these browser tasks built in.

ps. Is it not called Robotic Process Automation? First time I'm hearing it as Remote process Automation.

[1]https://github.com/ProductLoft/arachne

[2]https://www.skyvern.com/

[3]https://github.com/reworkd/tarsier