Lightnews — Scholar-powered news

David Gasquez @davidgasquez.com · 6h

This pattern works like a charm with LLMs! Start with a simple and dumb question, get details, iterate it!

David Gasquez @davidgasquez.com · 2d

Las Lightning Talks suelen ser de mis momentos favoritos en las PyCon. Una pena que no haya este año. ¡El programa, por otro lado, se ve genial!

1

Reposted by David Gasquez

Simon Willison @simonwillison.net · 10d

I've become one of those people who run multiple coding agents at once - here are my notes on embracing the parallel coding agent lifestyle simonwillison.net/2025/Oct/5/p...

Embracing the parallel coding agent lifestyle

For a while now I’ve been hearing from engineers who run multiple coding agents at once—firing up several Claude Code or Codex CLI instances at the same time, sometimes in …

simonwillison.net

25 14 160

David Gasquez @davidgasquez.com · 9d

Love teaching teaching agents about CLIs!

davidgasquez.com/cursor-agent...

It's a bit dated now (write it on January 2025!) but showing them `uvx` and `npx` is an amazing way to give tons of CLIs without worring what is installed or not.

David Gasquez

David Gasquez personal website

davidgasquez.com

1

Reposted by David Gasquez

Gergely Orosz @gergely.pragmaticengineer.com · 13d

When you try to predict about the impact of AI tools on software engineering - a great analogy I heard from @simonwillison.net is to look at the impact of smartphones (+ the amazing cameras they have) on filmmaking.

Noticed that those able to shoot pro films using a smartphone are... pros?

4 11 96

Reposted by David Gasquez

Paul Frazee @pfrazee.com · Apr 24

I personally see collapse as an inevitability for every individual institution. Nothing lasts forever. The smart way to build systems is to enable reforms so that they can adjust to reality without destroying what works in the process.

This is pretty baked into the philosophy of Bluesky

3 17 140

Reposted by David Gasquez

Gordon @gordon.bsky.social · Sep 15

if only there were a decentralized, immutable, and versioned way to fix this!

Conrad Hackett @conradhackett.bsky.social · Sep 14

LINK ROT: 38% webpages that existed in 2013 were no longer available 10 years later.

Even among pages that existed in 2021, 22% no longer accessible just two years later. This is often because individual page was deleted or removed on otherwise functional website.

Many implications for knowledge 🧪

A line chart showing that 38% of webpages from 2013 were no longer accessible one decade later.

1 1 7

Reposted by David Gasquez

Shawn Simister @narphorium.com · Aug 6

I wrote up some thoughts about why we’re seeing a gap in AI tool use among different types of developers
narphorium.com/blog/top-dow...
1/4

Top-Down vs. Bottom-Up Development | Shawn Simister

Why do AI tools work for some developers but not others?

narphorium.com

3 1 5

David Gasquez @davidgasquez.com · Sep 5

Oh, love this. Thanks for sharing Shawn!

1

David Gasquez @davidgasquez.com · Sep 4

I used to get to the same state ("feeling" how I want it to be done) by researching and looking at similar problems. Now, I go directly to implementation and learn by watching agents do it.

2

David Gasquez @davidgasquez.com · Sep 4

Is there a name for the pattern where you try building the same thing (feature, project, ...) a bunch of times with different agents to get a sense of the implementation design space?

I usually run things at least 2 or 3 times until I "feel" what is the way I want it to be built.

4 1 5

David Gasquez @davidgasquez.com · Aug 27

Been following the new notebook approach and really like it! Awesome work, Mike!

2

David Gasquez @davidgasquez.com · Aug 27

Both Evidence and Observable Framework seem to be slowing down in development.

- github.com/observablehq...
- github.com/evidence-dev...

Not sure why. Anyone has any context or ideas?

1 1 1

David Gasquez @davidgasquez.com · Aug 12

Setting up @tangled.sh as a mirror for a GitHub repository is simple:

$ git remote set-url --add --push origin [email protected]:TANGLED_USERNAME/REPOSITORY

$ git remote set-url --add --push origin [email protected]:GH_USERNAME/REPOSITORY.git

1

David Gasquez @davidgasquez.com · Aug 12

My dotfiles repository is now mirrored on @tangled.sh too!

tangled.sh/@davidgasque...

1 4

David Gasquez @davidgasquez.com · Aug 8

Spent some time refining the approach and benchmarking it. Just published a blog post about the core idea.

Can a bunch of LLM Agents be used to rank an arbitrary set of items in a consistent way? 🤖

davidgasquez.com/ranking-with...

1 4

David Gasquez @davidgasquez.com · Aug 7

I can't compare against MacOS since I've never had one but I haven't had any major issues in my development laptop in years.🤞

I've had a couple of interesting debugging sessions on my desktop computer though! Too much tinkering without reading the consequences.

1

David Gasquez @davidgasquez.com · Aug 7

Thanks for sharing the experience! Linux is not perfect but is fun*! 😅

* If you are into this type of fun.

1 3

David Gasquez @davidgasquez.com · Aug 6

From hallucination to reality! Cool example! 😅

1 1

David Gasquez @davidgasquez.com · Aug 6

I am! The times I've used it (low risk one off tasks), it has worked pretty well.

In this case, I manually reviewed the summaries and they all seemed correct. I already had some context on the projects so it only took a quick check to verify.

1

David Gasquez @davidgasquez.com · Aug 6

In the old times, I would have written an ad hoc Python script using a random library to do the processing/cleaning.

These days, inlining Claude Code (or `llm`) is just so much easier and faster!

1

David Gasquez @davidgasquez.com · Aug 6

Wrote a post about a small pattern I've been using for a while. Replacing chunks of bash scripts with `llm` or inline Claude Code calls.

For example, for data cleaning/structuring.

davidgasquez.com/scrappy-data...

While working on a small analysis over highly unstructed data at work, I came up with a very hacky but effective and cheap way to clean and process messy data. There is almost no code involved. The main downside is that requires a Claude Code subscription for it to be cost effective!

davidgasquez.com

1 6

Reposted by David Gasquez

Gordon @gordon.bsky.social · Jul 29

usually the baseline is “nothing” so the right heuristic is usually not “is it ideal” but rather “is it better than nothing?”

1 1 8

David Gasquez @davidgasquez.com · Jul 25

Are there any benchmarks/evals that track model performance over time?

I've heard things like Claude/Cursor getting dummier over time (haven't noticed much myself) and am wondering if there is some data to back this.

1 2

David Gasquez @davidgasquez.com · Jul 24

Totally! 💯

Hopefully some of these learnings can be applied to Parquet too?

1 3