frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

A semantic POP-style framework for structuring AI-assisted programs

https://github.com/dohuyhoang93/theus/blob/main/README.md
3•dohuyhoangvn93•2h ago

Comments

dohuyhoangvn93•2h ago
Hi HN,

I'm the author of THEUS.

This work comes from experimenting with AI-assisted coding and noticing that many existing architectural patterns (Clean Architecture, DDD) assume a human-only authoring process.

THEUS explores a POP (Pipeline-Oriented Programming) model where each unit explicitly declares: - input - output - side effects - error semantics

The goal is not a framework with runtime magic, but a structural discipline that makes programs easier to reason about, compress, and modify — for both humans and AI tools.

This is still exploratory. I'm mainly looking for: - prior art I may have missed - failure modes of this approach - cases where this abstraction clearly does not help

Happy to discuss.

For those interested in a more formal description, there's a DOI-linked whitepaper here: https://doi.org/10.5281/zenodo.18016750

PipeWire Guide

https://github.com/mikeroyal/PipeWire-Guide
1•vlr•28s ago•0 comments

Essential Education ChatGPT Prompts for Best Studying Practices

https://tools.eq4c.com/ai-prompts/10-essential-education-chatgpt-prompts-for-best-studying-practi...
1•edures•1m ago•0 comments

Coursera to acquire Udemy to create $2.5B MOOC giant

https://www.highereddive.com/news/coursera-to-acquire-udemy-to-create-25b-mooc-giant/808212/
1•pknerd•3m ago•0 comments

Dutch Tesla Fleet Goes Bankrupt After Betting on Musk's Self-Driving Promises

https://guessingheadlights.com/dutch-tesla-fleet-goes-bust-after-betting-on-musks-self-driving-pr...
1•toss1•3m ago•0 comments

Hiring a VPE/CTO for Startups (Part 1)

https://shailppatel.com/2025/12/23/essential-guide-to-hiring-a-vpe-cto-for-startups-part-1/
1•sna1l•7m ago•0 comments

Philosopher of Pride

https://aeon.co/essays/the-hidden-role-of-pride-and-shame-in-the-human-hive
1•benbreen•9m ago•0 comments

Schrödinger's test: The /dev/mem case

https://elisa.tech/ambassadors/2025/12/10/schrodingers-test-the-dev-mem-case/
1•signa11•11m ago•0 comments

Maincoder-1B – an open 1B-parameter coding model with 76% HumanEval

https://huggingface.co/Maincode/Maincoder-1B
1•MainNews•14m ago•1 comments

Show HN: Dead-simple publishing for Markdown from your automation flows

https://md2.website/
1•sotoplatero•15m ago•1 comments

Attention from First Principles

https://metaworld.me/blog/public/Attention-From-First-Principles
1•todsacerdoti•17m ago•0 comments

Xgboost

https://github.com/dmlc/xgboost
1•handfuloflight•23m ago•0 comments

An initial analysis of the discovered Unix V4 tape

https://www.spinellis.gr/blog/20251223/?ms261223
1•MBCook•23m ago•0 comments

On Friday Deploys: Sometimes That Puppy Needs Murdering

https://substack.com/home/post/p-182465670
1•gpi•24m ago•0 comments

LSP, Hooks, and Workflow Design: What Differentiates AI Coding Tools

https://blog.dataengineerthings.org/lsp-hooks-and-workflow-design-what-actually-differentiates-ai...
2•chtefi•29m ago•1 comments

Why HTTP-based evals worked better for our AI team than SDK-only setups

https://www.getmaxim.ai/blog/beyond-the-sdk-why-ai-teams-love-http-endpoint-based-evals/
1•PranayBatta•32m ago•0 comments

Mathematics Training Options

https://old.reddit.com/r/academiceconomics/comments/1hyljpm/mathematics_training_options/
1•barry-cotter•36m ago•0 comments

Salesforce pulls back from large language models after reliability concerns

https://www.moneycontrol.com/europe/?url=https://www.moneycontrol.com/technology/salesforce-pulls...
2•walterbell•37m ago•0 comments

Human Capital, Not "Industrial Policy," Explains East Asian Success

https://www.richardhanania.com/p/human-capital-not-industrial-policy
1•gmays•40m ago•0 comments

-tucky

https://languagelog.ldc.upenn.edu/nll/?p=58650
2•benatkin•47m ago•0 comments

PlugOS – An isolated OS that runs on an untrusted phone

1•drding•55m ago•1 comments

Show HN: Open-source Planning Poker web app with no voting limits

https://github.com/rie03p/planning-poker
1•rie03p•59m ago•0 comments

Do we need semantic layer anymore? What are the limitations of LLMs?

https://motherduck.com/blog/who-needs-a-semantic-layer-anyway/
1•thamizhan2611•1h ago•0 comments

Björk does calm Teardown of Sony Trinitron TV, Christmas 1988 [video]

https://www.youtube.com/watch?v=SNQtQWjX-sA
4•keepamovin•1h ago•0 comments

Dolphin Progress Release 2512

https://dolphin-emu.org/blog/2025/12/22/dolphin-progress-report-release-2512/
1•zdw•1h ago•0 comments

Old English Computer Glossary

https://web.archive.org/web/20231120210517/http://www.u.arizona.edu/~ctb/wordhord.html
2•LAC-Tech•1h ago•0 comments

Claude Code with API Key?

https://old.reddit.com/r/ClaudeAI/comments/1jwvssa/comment/mtt0urz/
1•behnamoh•1h ago•0 comments

Microsoft wants to replace its C and C++ codebase, perhaps by 2030

https://www.theregister.com/2025/12/24/microsoft_rust_codebase_migration/
1•0in•1h ago•1 comments

Pennsylvania High Court Rules Police Can Access Google Searches Without Warrant

https://reclaimthenet.org/pennsylvania-court-rules-no-privacy-in-google-searches
2•imglorp•1h ago•1 comments

A new immunotherapy approach could work for many types of cancer

https://news.mit.edu/2025/new-immunotherapy-approach-could-work-many-types-cancer-1216
2•0in•1h ago•0 comments

QWED – Deterministic Verification for AI

https://docs.qwedai.com/
1•handfuloflight•1h ago•0 comments