frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

The Agentic Systems Series

https://gerred.github.io/building-an-agentic-system/index.html
1•vinhnx•2m ago•0 comments

Ask HN: How do you stop teams from overestimating sprint capacity?

1•harryosgood•2m ago•0 comments

Victoria Woodhull – The first woman to run for president

https://en.wikipedia.org/wiki/Victoria_Woodhull
1•leopoldj•7m ago•0 comments

Gates-Backed Study: Flu Shots Linked to 27% Higher Heart Injury Risk

https://modernity.news/2025/09/04/gates-backed-study-flu-shots-linked-to-27-higher-heart-injury-r...
1•bilsbie•9m ago•1 comments

996

https://lucumr.pocoo.org/2025/9/4/996/
2•genericlemon24•9m ago•0 comments

Scale AI former CTO launches AI agent that could solve big data biggest problem

https://techcrunch.com/2025/09/05/scale-ais-former-cto-launches-ai-agent-that-could-solve-big-dat...
1•jimexp69•11m ago•0 comments

Ask HN: Why do LLMs struggle with word count?

1•rishikeshs•12m ago•0 comments

A16Z scouting ambitious Swiss founders for $1M accelerator

1•boggsss•13m ago•0 comments

Do Hangover Supplements Work?

https://www.economist.com/science-and-technology/2025/09/05/do-hangover-supplements-work
1•Anon84•14m ago•1 comments

The Asshole Filter

https://mrsteinberg.com/the-asshole-filter/
1•317070•14m ago•0 comments

Lidar can destroy phone camera sensors [video]

https://www.youtube.com/shorts/oeHtfMFdzIY
1•eproxus•15m ago•0 comments

'So Here's This Lunatic with a Crew Cut'

https://www.vulture.com/article/interview-snl-band-leader-ge-smith-best-songs-of-his-career.html
1•PaulHoule•18m ago•0 comments

US jobs market weakens further in August

https://www.bbc.com/news/articles/cn0xe5dvp47o
2•belter•19m ago•0 comments

In 1954, an Extraterrestrial Bruiser Shocked This Alabama Woman

https://www.smithsonianmag.com/smithsonian-institution/1954-extraterrestrial-bruiser-shocked-alab...
1•m-hodges•19m ago•0 comments

A Thornton high school teacher helped Colorado choose official state mushroom

https://coloradosun.com/2025/09/06/greg-sanchez-students-state-mushroom/
1•mooreds•19m ago•0 comments

Show HN: Pompelmi – Client-side file upload pre-quarantine (no cloud)

https://github.com/pompelmi/pompelmi
1•alrjoe•19m ago•0 comments

Managing NPM Releases

https://pilcrowonpaper.com/blog/managing-npm-releases/
1•mooreds•20m ago•0 comments

PHP MCP SDK: official client/server framework for MCP

https://github.com/modelcontextprotocol/php-sdk
2•mustaphah•21m ago•0 comments

Engineering Excellence Starts on Edge

https://world.hey.com/dhh/engineering-excellence-starts-on-edge-c36e4c59
1•ksec•24m ago•0 comments

Patterns vs. "Patterns"

https://perl.plover.com/classes/design/samples/slide012a.html
1•todsacerdoti•24m ago•0 comments

We Hacked Burger King: How Auth Bypass Led to Drive-Thru Audio Surveillance

https://bobdahacker.com/blog/rbi-hacked-drive-thrus/
4•BobDaHacker•25m ago•2 comments

GOP may succeed in unrelenting quest to kill two NASA climate satellites

https://arstechnica.com/space/2025/09/gop-may-finally-succeed-in-unrelenting-quest-to-kill-two-na...
3•throw0101c•30m ago•1 comments

My 14 year old brother discovered this interesting and repeatable GPT failure

https://xcancel.com/arithmoquine/status/1964179963323830624
2•Alifatisk•32m ago•0 comments

AI and the Future of Governance, Risk, and Compliance (GRC)

https://trainingcamp.com/articles/ai-and-the-future-of-governance-risk-and-compliance-grc/
1•joeldox•33m ago•0 comments

Mouse-Free Setups

https://lobste.rs/s/o0x7rb/your_mouse_free_setups
1•simonpure•37m ago•0 comments

Write Your Own Retro Compiler

https://t3x.org/t3x/0/book.html
2•AlexeyBrin•38m ago•1 comments

Ask HN: How do you turn engineering metrics into help and not blame?

1•harryosgood•39m ago•1 comments

OpenAI reorg at risk as Attorneys General push AI safety

https://www.theregister.com/2025/09/05/openai_reorg_at_risk/
3•rntn•46m ago•0 comments

Show HN: QuickDeploy – Deploy your web app to a VPS with one command

https://quickdeploy.dev
2•programonaut•51m ago•2 comments

Hirschmilch: Since 2008 free electronic music 24/7

https://hirschmilch.de/homepage
2•fbn79•55m ago•0 comments
Open in hackernews

A Software Development Methodology for Disciplined LLM Collaboration

https://github.com/Varietyz/Disciplined-AI-Software-Development
15•jay-baleine•2h ago

Comments

sublinear•32m ago
This may produce some successes, but it's so much more work than just writing the code yourself that it's pointless. This structured way of working with generative AI is so strict that there is no scaling it up either. It feels like years since this was established to be a waste of time.

If the goal is to start writing code not knowing much, it may be a good way to learn how and establish a similar discipline within yourself to tackle projects? I think there's been research that training wheels don't work either though. Whatever works and gets people learning to write code for real can't be bad, right?

weego•19m ago
It's just a function of how much code you need to write, and how much un-interrupted time you have.

Editing this kind of configuration has far less cognitive load and loading time, so distractions aren't as destructive to the task as they are when coding. You can then also structure time so that productive agent coding can be happening while you're doing business critical tasks like meetings / calls etc.

I do think this is overkill though, and it's a bad plan and far too early to try and formalize The One Way To Instruct AI How To Code, but every advance is an opportunity to gain career traction so fair play.

CuriouslyC•7m ago
The most important thing is to have a strong plan cycle in front of you agent work, if you do that, agents are very reliable. You need to have a deep research cycle that basically collects a covering set of code that might need to be modified for a feature, feeds it into gemini/gpt5 to get a broad codebase level understanding, then has a debate cycle on how to address it, with the final artifact being a hyper detailed plan that goes file by file and provides an outline of changes required.

Beyond this, you need to maintain good test coverage, and you need to have agents red-team your tests aggressively to make sure they're robust.

If you implement these two steps your agent performance will skyrocket. The planning phase will produce plans that claude can iterate on for 3+ hours in some cases, if you tell it to complete the entire task in one shot, and the robust test validation / change set analysis will catch agents solving an easier problem because they got frustrated or not following directions.

esafak•5m ago
Modern agent tools already draw up plans before implementation. Some of them even define "plan" and "build" agents: https://opencode.ai/docs/agents/#built-in