Hacker News Reader

Really cool and interactive. Sharing this with my kids in high school.

taintechon A Design Space Exploration of Async/Await

Really nice paper, it would be great to add Scala too.

The fire-and-forget log write is a great choice of example.

taintechon Nobody Asked AI to Hack Hugging Face. So Why Did It?

Motive, it is interesting angle to take. I would also recommend reading “ Revealing the details of how OpenAI agents hacked Hugging Face” https://swarmtraces.org/

taintechon California Attorney General Serves Investigative Subpoena on OpenAI

I think it is quite a routine for OpenAI right now.

taintechon Why you can need Cloud Run Instances

Easy pattern in Cloud Run, if anyone is interested for AI agents.

Use invocation based Cloud Run with Cloud Tasks.

User calls service in Cloud Run, Cloud Run start Cloud Task that holds connection for 30 minutes. Cheap and easy patterns, and you avoid no cpu throttling too.

I use it with ADK agents, works well.

taintechon Improving Throughput by Optimising KV Cache Efficiency for Agentic Workloads

I really like this article, agent harness knows its own state: which tool call is running, whether a sub-agent is pending, when it is finished. The API throws that away because each request looks stateless. Interesting idea.

taintechon Real-SWE: Benchmarking AI models on private, real-world, enterprise codebases

I think openrca https://github.com/microsoft/OpenRCA is quite interesting

taintechon Real-SWE: Benchmarking AI models on private, real-world, enterprise codebases

Thank you sir!

taintechon Real-SWE: Benchmarking AI models on private, real-world, enterprise codebases

I think currently AI models did incredible improvement against 2025 OpenRCA research paper with 11.25% success rate.

Key questions, are they topped in performance? Is there some next leap?

taintechon LLMs as Retrieval and Recommendation Engines

This is a cool idea, but the cost is a killer. Normally, you would run recommendations and pre-cache them for the users or items they are associated with. Running a giant LLM for every user's recommendation is thousands of times more expensive and slower than current methods. It just doesn't seem practical for a large number of users.

taintechon Big clouds scramble as EU Data Act brings new data transfer rules

Nice to see Google pushing zero-cost transfers, hope others will follow.

taintechon What if the $3T AI investment boom goes wrong?

World changing by AI is real, but the impact will be different than anyone expects.

Big tech has huge FOMO and is over-investing, securing their advantage. This will cause a crash, but the crash will create a lot of new opportunities for small players, in my opinion.