Hacker News

gpt-5.3-codex is finally available on the API!

Comments URL: https://news.ycombinator.com/item?id=47143356

Points: 1

# Comments: 0

Article URL: https://www.amazon.co.uk/dp/B0GHT81F1K

Comments URL: https://news.ycombinator.com/item?id=47143349

Points: 2

# Comments: 0

Article URL: https://seekingalpha.com/news/4555907-ai-chip-startup-matx-raises-500m-in-race-to-compete-with-nvidia

Comments URL: https://news.ycombinator.com/item?id=47143339

Points: 1

# Comments: 0

Article URL: https://www.theclimatebrink.com/p/the-ai-augmented-scientist

Comments URL: https://news.ycombinator.com/item?id=47143333

Points: 1

# Comments: 0

Article URL: https://blog.thalesouza.com.br/en/blog/logic-migration/

Comments URL: https://news.ycombinator.com/item?id=47143331

Points: 1

# Comments: 1

We built an open-source pipeline that scrapes HN's top stories daily, generates summaries with LLMs, and publishes a podcast on Spotify. Perfect for commutes or gym sessions when you can't read threads.

The pipeline: scrape HN -> LLM summarization -> script generation -> TTS -> auto-publish to Spotify.

Completely free, no ads. We're not affiliated with HN or YC - just longtime readers who wanted a companion tool. For the real discussions, nothing beats the HN comment threads.

Would love feedback from the community that inspired this project.

Comments URL: https://news.ycombinator.com/item?id=47143324

Points: 1

# Comments: 0

Article URL: https://www.nytimes.com/2026/02/24/opinion/ezra-klein-podcast-jack-clark.html

Comments URL: https://news.ycombinator.com/item?id=47143289

Points: 1

# Comments: 0

Article URL: https://www.nytimes.com/athletic/7058361/2026/02/20/nba-tanking-how-to-fix-rule-changes/

Comments URL: https://news.ycombinator.com/item?id=47143285

Points: 1

# Comments: 0

Article URL: https://vlaaad.github.io/clojure-survey-trends

Comments URL: https://news.ycombinator.com/item?id=47143284

Points: 1

# Comments: 0

Article URL: https://wonderwhy-er.medium.com/mcp-apps-ai-just-got-a-frontend-and-we-built-one-cd1f6fc762f6

Comments URL: https://news.ycombinator.com/item?id=47143273

Points: 1

# Comments: 1

Article URL: https://news.crunchbase.com/venture/data-2025-vs-2021-funding-hottest-companies-ai/

Comments URL: https://news.ycombinator.com/item?id=47143272

Points: 1

# Comments: 0

Article URL: https://ai-tools-woad-six.vercel.app

Comments URL: https://news.ycombinator.com/item?id=47143267

Points: 1

# Comments: 0

Article URL: https://www.usatoday.com/story/entertainment/celebrities/2026/02/24/robert-carradine-death-lizzie-mcguire-actor-revenge-nerds/88838510007/

Comments URL: https://news.ycombinator.com/item?id=47143254

Points: 1

# Comments: 1

Article URL: https://mercury.com/blog/startups-ai-icp

Comments URL: https://news.ycombinator.com/item?id=47143250

Points: 1

# Comments: 0

Article URL: https://terallite.substack.com/p/the-true-face-of-prompt-injection

Comments URL: https://news.ycombinator.com/item?id=47143247

Points: 1

# Comments: 0

So, I have been investing for some time now. At first, it was easier to track my portfolio because I was just using one broker, but as time went on, my portfolio became bigger and more diversified (for example, I bought physical gold, deposited money into a savings account at Bank A, and also used Broker A and Broker B because only Broker B had the stocks I wanted to buy).

It just got harder to track my investments, it was also getting harder to understand whether my investments were aligned with the portfolio I had built.

So naturally, I started to search for solutions. At first, I found a few desktop and mobile apps. But the problem with the majority of them was that they were either too complicated to use or just over-engineered. Nearly all the apps had a FIRE calculator, were synced with the market (which was logical, but how can I get the price for physical gold?), or were also trying to track my expenses. I just wanted to track my portfolio. Hence came the second option: using Excel.

And actually, this is the way the majority of people do it. Knowing that, I tried to create an Excel sheet for myself. But the barrier to entry was just too high; I didn't know how to use it, so it just seemed too hard to implement a solution for myself. Also, the user experience just didn't feel very good. I wanted to see pie charts, good fonts, etc. (I could probably do these things with Excel as well, but if I can't even implement a simple sheet, how could I do these cool visuals?).

So, I decided to implement my own solution. My needs were really simple: - I want to see all my investments on one screen. - I want to see my P&L. - I want to see whether my investment ratio is aligned with my portfolio. - I want to sync it across different devices. - I want to have different currencies (like EUR, TL) because I invest in different markets. - I want it to be free. - I want to see how my investments grow over time.

And that's about it. So, keeping all these things in mind, I built a web app for myself and wanted to share it with you

Comments URL: https://news.ycombinator.com/item?id=47143238

Points: 1

# Comments: 0

Article URL: https://github.com/spranab/brainstorm-mcp

Comments URL: https://news.ycombinator.com/item?id=47143222

Points: 1

# Comments: 1

Article URL: https://www.math.columbia.edu/~woit/wordpress/?p=15500

Comments URL: https://news.ycombinator.com/item?id=47143211

Points: 11

# Comments: 0

Hi HN,

Most of us have been there: It’s 3 AM, there’s an outage, and the #incident channel is exploding with 200+ messages. Once the fix is deployed, the real pain begins—spending 4 hours reconstructing the timeline for the post-mortem.

I built ProdRescue AI to automate this. It’s an incident intelligence engine that correlates technical logs with human context from Slack.

How it works:

Native Slack Integration: Connect via OAuth 2.0. We only access channels you explicitly invite the bot to.

Contextual Correlation: It maps Slack timestamps to log events, identifying not just what failed, but who made which decision and why.

4-Layer Intelligence: We use a pipeline to Sanitize (mask PII), Correlate (logs + chat), Infer (RCA), and Verify (link every claim to a source log line).

Security: We use ephemeral processing. No log retention, no training on your data.

I’m really interested in your thoughts on the "Evidence-Backed" approach. Instead of just generating a narrative, we link every finding to a specific evidence tag ([1], [2], etc.) to eliminate AI hallucinations.

Check it out here: https://prodrescueai.com

Would love to hear your feedback on the Slack-to-Timeline flow!

Comments URL: https://news.ycombinator.com/item?id=47142521

Points: 1

# Comments: 0

Article URL: https://shipordie.club/roast/startup

Comments URL: https://news.ycombinator.com/item?id=47142502

Points: 1

# Comments: 0