Project ideas from Hacker News discussions.

NAND-16: a computer built from 277,248 NAND gates

📝 Discussion Summary (Click to expand)

Three prevalent themes in the discussion


1. AI‑generated retro projects feel less impressive

“This is pretty sad. Seeing this a couple of years ago would be mind blowing. An impressive technical feat by someone with grit and resilience to painstakingly build this from scratch. Today? Meh.” — guiambros
“I got excited for a bit, until I loaded the page and saw it was slop. I still would be impressed if someone actually took the time to learn about and implement it for themselves.” — Retr0id

2. Concerns about authorship and proof when using LLMs

“there's also no way to prove I created it without LLM help” — LarsDu88
“You don’t need to prove it, that was never a requirement. You can simply say it. The only complication is that there’s a fair number of people who … really want to pass LLM output as their own.” — socializer
“Of course there is — ask you about it…. No way to succeed there without actually understanding.” — dmitrygr

3. The lasting value of hands‑on work for learning, reputation, and community

“I’m very grateful I have public pre‑LLM projects I can point at (including a 6502 emulator!), and I have absolutely no idea how I’d go about bootstrapping the same kind of reputation starting from scratch.” — Retr0id
“What was the point before? To have fun, build a community, and to get attention of hiring managers. The first two reasons are still there.” — socializer
“I think the takeaway is that we have to think bigger, much bigger, now that AI coding is essentially ‘solved.’” — aabajian


🚀 Project Ideas

CodeCred: Human‑Coding Verification Platform

Summary

  • Records IDE activity to produce tamper‑evident proof that code was written manually (or with limited AI assistance).
  • Enables developers to showcase authentic work to employers and peers.

Details

Key Value
Target Audience Developers, students, job seekers who need to prove coding authenticity
Core Feature IDE plugin that logs keystrokes, edits, timestamps; generates cryptographic receipt and optional zero‑knowledge proof of human effort
Tech Stack VS Code/IntelliJ extension (TypeScript), Rust backend for signatures, IPFS/Filecoin storage, circom/ZKP library
Difficulty Medium
Monetization Revenue-ready: subscription tiers (free basic, $9/mo Pro for unlimited proofs and verification badge)

Notes

  • HN commenters lament the inability to prove manual work (LarsDu88: “there's also no way to prove I created it without LLM help”) and value reputation (socializer: “reputation you build over time”).
  • Provides a concrete, verifiable way to earn that reputation, sparking discussion about trust in the AI‑era and offering practical utility for resumes, interviews, and open‑source contributions.

RetroBuild Showcase: Verified Hardware Projects

Summary

  • Platform for submitting retro computing projects with required build evidence (photos, video, logs).
  • Awards a “Human Built” badge after community verification.

Details

Key Value
Target Audience Retro computing hobbyists, educators, makers
Core Feature Verification workflow with evidence checklist (photos, timelapse video, BOM receipts) and reputation‑based badge system
Tech Stack Next.js (React/Tailwind), PostgreSQL, AWS S3 for media, optional GitHub Actions for HDL verification
Difficulty Medium
Monetization Hobby

Notes

  • Users express longing for tangible proof of pre‑LLM skill (Retr0id: “I have absolutely no idea how I'd go about bootstrapping the same kind of reputation starting from scratch”) and admiration for genuine builds (Nurysso: “OHH my god this is soo awesome”).
  • A trusted showcase would let HN readers share authentic retro projects, stimulate discussion about hardware craftsmanship, and give newcomers a clear path to earn credibility.

LearnByDoing: Guided Deep‑Dive Coding Challenges

Summary

  • Structured, milestone‑based deep‑dive projects (e.g., build a CPU from NAND gates) that require manual implementation and explanation.
  • Integrates optional viva/video defense to certify understanding.

Details

Key Value
Target Audience Students, self‑learners, interview candidates preparing for technical roles
Core Feature Guided curriculum with checkpoints, auto‑graded tests, and required recorded explanation or live chat to verify comprehension
Tech Stack React frontend, Node.js/Express API, Docker sandbox for code execution, AWS MediaStore for viva recordings, Streamlit for interactive feedback
Difficulty High
Monetization Revenue-ready: course sales ($49 per track) or corporate licensing

Notes

  • Commenters note that interview‑style talks defeat LLMs (dmitrygr: “give a talk ... then answer questions … No way to succeed there without actually understanding”) and lament the loss of meaningful projects (aabajian: “we have to think bigger… now that AI coding is essentially ‘solved’”).
  • LearnByDoing offers a path to demonstrate deep understanding beyond just code, fostering discussion on effective learning in the AI era and providing a practical tool for job seekers and educators.

Read Later