adsharma

Joined 10 years ago

Karma

556

Submissions

434

Member since

Mar 19

ladybugdb.com

View on Hacker News

Recent submissions

Comment · 4 days ago

More interested in creating a variant of python that's practical.

Wrote a parser with a rust compatible pattern matching syntax here:

https://github.com/py2many/spy-ast

Should help both JIT and static translation.

Comment · 4 days ago

Grokipedia's mistake is that it generates text statically instead of dynamically from a machine curated data structure in latent space.

Latentpedia starts from wikidata, not Wikipedia.

Comment · 4 days ago

Latentpedia.org will not have these problems. The content is human produced, but the extraction of a graph and creating the latent space is AI driven.

Comment · 4 days ago

It's strange that typescript has 5 different LLM written competing static translators to compiled languages and python is killing JIT because the PEP was not accepted.

Comment · 4 days ago

This is largely the idea behind py2many + fquery, but we want to get rid of C-API as well.

  https://github.com/py2many/py2many
  https://github.com/adsharma/fquery/blob/main/tests/django_example_model.py

Comment · 8 days ago

Not sarcastic. Backed by a registered domain name, GitHub project, discussion links, huggingface dataset and an OSS graph database project with 1900 stars.

The extremists are the ones who believe that Wikipedia is working well and quote Poe's law as described on the page, which Google will cite literally.

edit: Longer form blog post response including Google AI analysis

https://adsharma.github.io/google-wikipedia-and-poes-law/

Comment · 11 days ago

What academics need to defend is wikipedia in latent space. Not walls of text that reflect the opinions of a few who have the time to edit that the rich and famous can optimize.

That is why we are building latentpedia.org.

Comment · 16 days ago

This line of work is a mirage. You're much better off investing in a project like latentpedia.org and building a new class of LLMs which support dual use. Deterministic and less compressed or probabilistic and fit in GPU memory.

Comment · 18 days ago

There is a strong human incentive to have a monorepo in a corporate environment. This is not a technical question.

An entire dept exists to make it work. Atomicity and security are other arguments. The larger the company, the stronger the incentive.

It's not something average software developers working elsewhere care about much.

Amazon seems to be the only big company bucking this trend.

Comment · 20 days ago

  uv pip install nx-arrow

Comment · 20 days ago

Qualcomm CEO was talking demand for chips that can run a 100B param model on a phone.

There is a Q about how much RAM these phones will need and the average selling price.

People are seriously addicted to doom scrolling. But what are they willing to pay for a phone?

Comment · 25 days ago

We raised X million to kill Y

Is all over LinkedIn and Twitter

Comment · 26 days ago

LBUG_DUMP_LOGICAL=1 <run your query>

Improvements welcome!

Comment · 26 days ago

Maintainer of LadybugDB here. I'm reviewing a PR to transpile GQL to Cypher.

It should be possible to transpile datalog or another query language that is interesting.

Implementing storage and secondary indicies is the hard part.

Won't comment on rewriting the DB in Rust beyond what's in GitHub discussions.