Hacker News

Article URL: https://www.wired.com/story/cbp-photo-everyone-in-vehicle-us-border/

Comments URL: https://news.ycombinator.com/item?id=43906436

Points: 1

# Comments: 1

Article URL: https://www.rockstargames.com/newswire/article/3928aaa9471o3a/grand-theft-auto-vi-watch-trailer-2-now

Comments URL: https://news.ycombinator.com/item?id=43906393

Points: 2

# Comments: 1

Article URL: https://arxiv.org/abs/2504.14064

Comments URL: https://news.ycombinator.com/item?id=43906379

Points: 1

# Comments: 0

Article URL: https://www.youtube.com/watch?v=IBzY06tH0sQ

Comments URL: https://news.ycombinator.com/item?id=43906365

Points: 1

# Comments: 0

Article URL: https://www.cloudquery.io/blog/how-we-handle-billion-row-clickhouse-inserts-with-uuid-range-bucketing

Comments URL: https://news.ycombinator.com/item?id=43906357

Points: 2

# Comments: 0

Article URL: https://github.com/Milkdown/milkdown

Comments URL: https://news.ycombinator.com/item?id=43906355

Points: 1

# Comments: 0

Hey HN! We’re Vaibhav and Marcello. We’re building Plexe (https://github.com/plexe-ai/plexe), an open-source agent that turns natural language task descriptions into trained ML models. Here’s a video walkthrough: https://www.youtube.com/watch?v=bUwCSglhcXY.

There are all kinds of uses for ML models that never get realized because the process of making them is messy and convoluted. You can spend months trying to find the data, clean it, experiment with models and deploy to production, only to find out that your project has been binned for taking so long. There are many tools for “automating” ML, but it still takes teams of ML experts to actually productionize something of value. And we can’t keep throwing LLMs at every ML problem. Why use a generic 10B parameter language model, if a logistic regression trained on your data could do the job better?

Our light-bulb moment was that we could use LLMs to generate task-specific ML models that would be trained on one’s own data. Thanks to the emergent reasoning ability of LLMs, it is now possible to create an agentic system that might automate most of the ML lifecycle.

A couple of months ago, we started developing a Python library that would let you define ML models on structured data using a description of the expected behaviour. Our initial implementation arranged potential solutions into a graph, using LLMs to write plans, implement them as code, and run the resulting training script. Using simple search algorithms, the system traversed the solution space to identify and package the best model.

However, we ran into several limitations, as the algorithm proved brittle under edge cases, and we kept having to put patches for every minor issue in the training process. We decided to rethink the approach, throw everything out, and rebuild the tool using an agentic approach prioritising generality and flexibility. What started as a single ML engineering agent turned into an agentic ML "team", with all experiments tracked and logged using MLFlow.

Our current implementation uses the smolagents library to define an agent hierarchy. We mapped the functionality of our previous implementation to a set of specialized agents, such as an “ML scientist” that proposes solution plans, and so on. Each agent has specialized tools, instructions, and prompt templates. To facilitate cross-agent communication, we implemented a shared memory that enables objects (datasets, code snippets, etc) to be passed across agents indirectly by referencing keys in a registry. You can find a detailed write-up on how it works here: https://github.com/plexe-ai/plexe/blob/main/docs/architectur...

Plexe’s early release is focused on predictive problems over structured data, and can be used to build models such as forecasting player injury risk in high-intensity sports, product recommendations for an e-commerce marketplace, or predicting technical indicators for algorithmic trading. Here are some examples to get you started: https://github.com/plexe-ai/plexe/tree/main/examples

To get it working on your data, you can dump any CSV, parquet, etc and Plexe uses what it needs from your dataset to figure out what features it should use. In the open-source tool, it only supports adding files right now but in our platform version, we'll have support for integrating with Postgres where it pulls all available data based on an SQL query and dumps it into a parquet file for the agent to build models.

Next up, we’ll be tackling more of the ML project lifecycle: we’re currently working on adding a “feature engineering agent” that focuses on the complex data transformations that are often required for data to be ready for model training. If you're interested, check Plexe out and let us know your thoughts!

Comments URL: https://news.ycombinator.com/item?id=43906346

Points: 7

# Comments: 1

Free to download, fully SVG files. You can use as a logo or placeholders, no attribution required.

Comments URL: https://news.ycombinator.com/item?id=43906345

Points: 1

# Comments: 1

Article URL: https://starcounter.vercel.app/

Comments URL: https://news.ycombinator.com/item?id=43906341

Points: 1

# Comments: 1

Article URL: https://www.ft.com/content/f5797d16-6299-444c-b990-5cbc6a7ccfca

Comments URL: https://news.ycombinator.com/item?id=43906340

Points: 1

# Comments: 0

Article URL: https://arxiv.org/abs/2504.17834

Comments URL: https://news.ycombinator.com/item?id=43905792

Points: 2

# Comments: 0

Article URL: https://www.utubepublisher.in/2025/05/Ancient-toilet-systems.html

Comments URL: https://news.ycombinator.com/item?id=43905790

Points: 2

# Comments: 0

Real Jacob here. This was a really weird experience, primarily hearing my own voice read words I wrote.

The newsletter is the sort of.. offboarding. The podcast episode is embeded, I hope you enjoy it, and that it's listenable!

I'm trying to untangle mine and other creators feelings on the paradox of AI: it's a threat and an opportunity. So what do we do about it?

Comments URL: https://news.ycombinator.com/item?id=43905777

Points: 1

# Comments: 0

Hi HN,

I've been experimenting for some time with writing web frameworks, and I thought it's time to publish one that I really like: it just embeds SQL directly inside HTML without any glue language.

One of the main motivators for getting this minimal was the Hundred Year Web Seervice with htmx presentation by Alexandrer Petros (https://www.youtube.com/watch?v=lASLZ9TgXyc), that showed that both HTML and SQL have staying power, but he couldn't really find a perfect glue language for his app. After some searching I only found ColdFusion to have implemented the same idea, but instead of the reactive direction that I want to go to, it went towards imperative programming which is not the direction I want to go to. I tried to learn from the original (pre-Adobe) ColdFusion, but make it closer to standard SQL.

I plan to keep the language super simple, just like a template language, make it provide 80% of the CRUD functionality a small web-site needs and add more extension points around it to handle other complexity by other languages.

I have already created a reactive SQL implementation, but the framework got too complex to use, so this is my next experiment (and HTMX can already provide a lot of interactivity).

Comments URL: https://news.ycombinator.com/item?id=43905771

Points: 1

# Comments: 0

More info here: https://forums.docker.com/t/i-want-to-monitor-internal-docke...

Comments URL: https://news.ycombinator.com/item?id=43905768

Points: 1

# Comments: 0

Article URL: https://samwho.dev/hashing/

Comments URL: https://news.ycombinator.com/item?id=43905766

Points: 1

# Comments: 0

Article URL: https://determinate.systems/posts/changelog-determinate-nix-342/

Comments URL: https://news.ycombinator.com/item?id=43905758

Points: 1

# Comments: 0

Article URL: https://github.com/arnarg/nixidy

Comments URL: https://news.ycombinator.com/item?id=43905751

Points: 1

# Comments: 0

Article URL: https://arstechnica.com/information-technology/2025/05/trumps-attacks-on-green-energy-are-big-trouble-for-data-centers-ai/

Comments URL: https://news.ycombinator.com/item?id=43905744

Points: 1

# Comments: 0

Article URL: https://www.mooncake.dev/blog/htap-is-dead

Comments URL: https://news.ycombinator.com/item?id=43905742

Points: 1

# Comments: 0