Hacker News: nvmdbljstm

New comment by nvmdbljstm in "Mistral 4 had a release, makes it comparable to Chinese open-weight models"

nvmdbljstm — Fri, 30 Oct 2026 09:27:45 +0000

~3x cheaper and ~3x quicker

Mistral 4 had a release, makes it comparable to Chinese open-weight models

nvmdbljstm — Fri, 30 Oct 2026 09:27:45 +0000

Article URL: https://www.reuters.com/world/china/mistral-ceo-says-new-ai-model-beats-chinese-ones-some-areas-2026-10-06/

Comments URL: https://news.ycombinator.com/item?id=50301537

Points: 3

# Comments: 1

Open Source or Custom in Enterprise AI Systems?

nvmdbljstm — Thu, 29 Oct 2026 10:56:07 +0000

Article URL: https://deepsense.ai/blog/when-code-gets-cheaper-judgment-gets-more-precious-quality-bottlenecks-in-enterprise-ai-systems/

Comments URL: https://news.ycombinator.com/item?id=50287093

Points: 4

# Comments: 0

Control Planes of Agentic AI Infrastructure for Reliable Software Delivery

nvmdbljstm — Thu, 29 Oct 2026 10:54:29 +0000

Article URL: https://deepsense.ai/blog/the-5-control-planes-of-agentic-ai-infrastructure-for-reliable-software-delivery/

Comments URL: https://news.ycombinator.com/item?id=50287073

Points: 3

# Comments: 0

New comment by nvmdbljstm in "The concept of wrapping your CLIs to make them more token-efficient [video]"

nvmdbljstm — Wed, 28 Oct 2026 14:10:19 +0000

Good point—though, tbh I’ve never benchmarked this myself. I just looked at the RTK gain, and it seemed to do the job." That said, because of my specific sandbox configuration, I’ve added gh and git to my "excluded tools" list, so I’m in no position to compare it against gh-axi.

New comment by nvmdbljstm in "The concept of wrapping your CLIs to make them more token-efficient [video]"

nvmdbljstm — Wed, 28 Oct 2026 14:06:50 +0000

Here’s an example of gh-axi, which wraps the GitHub CLI and prepares the most relevant information for agents—so the LLM doesn’t have to filter it out itself. So simple! I haven’t yet formed an opinion on whether this is overengineering or a genuine way to save time and money; I also haven’t looked into how it compares to the cost savings provided by RTK.

The concept of wrapping your CLIs to make them more token-efficient [video]

nvmdbljstm — Wed, 28 Oct 2026 14:05:56 +0000

Article URL: https://www.youtube.com/watch?v=TlmTypTQFj8

Comments URL: https://news.ycombinator.com/item?id=50273800

Points: 3

# Comments: 4

New comment by nvmdbljstm in "Ask HN: Has anyone used Phoenix, Langfuse, or OpenObserve?"

nvmdbljstm — Wed, 28 Oct 2026 08:58:31 +0000

Thanks a lot for your perspective. I needed a comparison cause I only used Phoenix. Its architecture is extremely simple. Tracing is nice, I didn't have any complaints. All other features like evaluations, datasets, prompts - I'd use a separate tool for this.

New comment by nvmdbljstm in "Astroterm: A Planetarium for Your Terminal"

nvmdbljstm — Wed, 28 Oct 2026 08:30:08 +0000

Regardless of whether it’s useful or not, the level of creative intelligence is impressive

Astroterm: A Planetarium for Your Terminal

nvmdbljstm — Wed, 28 Oct 2026 08:29:55 +0000

Article URL: https://github.com/da-luce/astroterm

Comments URL: https://news.ycombinator.com/item?id=50270651

Points: 155

# Comments: 14

Harness-aware evaluation of plugins (OpenAI Cookbook)

nvmdbljstm — Tue, 27 Oct 2026 14:32:08 +0000

Article URL: https://developers.openai.com/cookbook/examples/partners/harness_aware_plugin_evals/harness_aware_plugin_evals

Comments URL: https://news.ycombinator.com/item?id=50260007

Points: 4

# Comments: 0

Ask HN: Has anyone used Phoenix, Langfuse, or OpenObserve?

nvmdbljstm — Tue, 27 Oct 2026 10:44:41 +0000

What worked well, and what issues or limitations did you run into?

Comments URL: https://news.ycombinator.com/item?id=50257116

Points: 5

# Comments: 6

Ski buddy – no one knew they needed one, yet everyone did

nvmdbljstm — Mon, 26 Oct 2026 08:32:42 +0000

Article URL: https://chat.onthesnow.com/

Comments URL: https://news.ycombinator.com/item?id=50240812

Points: 3

# Comments: 0

LoRA can reduce GPU memory requirements from 42 GB to 7 GB [video]

nvmdbljstm — Fri, 23 Oct 2026 11:26:34 +0000

Article URL: https://www.youtube.com/watch?v=3lqcCs3jVdk

Comments URL: https://news.ycombinator.com/item?id=50207303

Points: 3

# Comments: 0

A framework for evaluating agentic AI systems

nvmdbljstm — Thu, 22 Oct 2026 07:32:51 +0000

Article URL: https://deepsense.ai/blog/your-agent-passed-the-eval-should-you-ship-it/

Comments URL: https://news.ycombinator.com/item?id=50192356

Points: 4

# Comments: 0

Anthropic wrote about replacing Git for their monorepo

nvmdbljstm — Wed, 21 Oct 2026 14:45:03 +0000

Article URL: https://www.anthropic.com/research/redesigning-source-control

Comments URL: https://news.ycombinator.com/item?id=50181921

Points: 4

# Comments: 1

New comment by nvmdbljstm in "AI Testing vs. Software Testing"

nvmdbljstm — Wed, 21 Oct 2026 12:21:12 +0000

How do the principles of traditional Quality Engineering apply to systems built on top of LLMs? In the article, I look at what it takes to move from a deterministic approach to a probabilistic one in modern AI delivery.

AI Testing vs. Software Testing

nvmdbljstm — Wed, 21 Oct 2026 12:20:22 +0000

Article URL: https://deepsense.ai/blog/ai-testing-vs-software-testing-how-to-build-reliable-production-ai-systems/

Comments URL: https://news.ycombinator.com/item?id=50179680

Points: 3

# Comments: 1

Show HN: A practical framework for evaluating agentic AI systems

nvmdbljstm — Tue, 20 Oct 2026 07:45:13 +0000

Article URL: https://deepsense.ai/blog/your-agent-passed-the-eval-should-you-ship-it/

Comments URL: https://news.ycombinator.com/item?id=50162634

Points: 3

# Comments: 0

New comment by nvmdbljstm in "Anyone Used the OAI Batch API or Azure OAI Batch API for Bulk Data Processing?"

nvmdbljstm — Tue, 20 Oct 2026 07:43:49 +0000

How was your experience with reliability, turnaround time, and cost efficiency? Any issues or lessons learned? How it compares to Claude Batch API?