Ask HN: How do you improve code for future AI?

1 tmaly 3 5/31/2025, 1:03:44 PM

I am sitting on tons of legacy code at my day job. I keep looking at how fast AI is improving over the last few years.

I am thinking what can I do to improve the massive amount of legacy code I have to maintain so that eventually I can let a better AI in the near future help me more?

What are your best tips that are working now?

Comments (3)

apothegm · 5h ago

AI has limited context windows and little ability to reason about what complex code does.

If you can refactor it to small modules with simple interfaces and properly document the interfaces and what the modules do, AI can work with your code more easily. Think of the ideal AI codebase as being assembled from libraries (even if private internally developed ones) with a small amount of glue code in between.

lucfranken · 14h ago

Did you try running some ai over it to see if it is capable of being useful?

Legacy code is a statement but in the end it only matters if it works and is maintainable.

It might surprise you.

tmaly · 5h ago

I got mixed results. A good portion of the code was written by former employees. There is little documentation.

FreeBSD 14.3-RC1 Now Available (lists.freebsd.org)

Why DeepSeek is cheap at scale but expensive to run locally (seangoedecke.com)

Famous Amiga Uses (wigilius.se)

Show HN: Discordz – A simple Discord server directory (discordz.com)

Show HN: I Made Notorium – AI Notes and Mind Maps from Live Lectures (notorium.app)

How to animate a mesh along the surface of a sphere (garden.bradwoods.io)

An optimizing compiler doesn't help much with long instruction dependencies (johnnysswlab.com)

Browser extension (Firefox, Chrome, Opera, Edge) to redirect URLs based on regex (github.com)

Manosphere (en.wikipedia.org)

Illinois Cops Gave ICE Access to More Than 5k Surveillance Cameras Nationwide (reason.com)

Takeaways from Jony Ive's Stripe Sessions Chat (spyglass.org)

How seaweed is a powerful, yet surprising, climate solution (theconversation.com)

Understanding Memory Management, Part 6: Basic Garbage Collection (educatedguesswork.org)

Building the Bucket Linear Agent (bucket.co)

ChatGPT: H1 2025 Strategy (justice.gov)

Solving Queuedle (healeycodes.com)

Show HN: Dungeon Newbie (an RPG game for total newbies) (rodyne.com)

Ports of Call (portsofcall.de)

Permuting Bits with GF2P8AFFINEQB (bitmath.blogspot.com)

A Beautiful Technique for Some XOR Related Problems (codeforces.com)

h2tunnel – TCP over HTTP/2 (boronine.com)

Google AI Edge – on-device cross-platform AI deployment (ai.google.dev)

The American vs. European Mindset on Life (mertbulan.com)

Create Popups, in Just 3 Clicks

I like to install NixOS (declaratively) (michael.stapelberg.ch)

Chomp (en.wikipedia.org)

Flatworms grow a second head instead of a tail, if influenced by electric field (newscientist.com)

Vintage Machinery (wiki.vintagemachinery.org)

How to Grow an LSM-Tree? Towards Bridging the Gap Between Theory and Practice (arxiv.org)

Kessel Run (kesselrun.af.mil)

Figma Slides Is a Beautiful Disaster (allenpike.com)

From Typewriters to Transformers: AI Is Just the Next Tools Abstraction (hardcoresoftware.learningbyshipping.com)

Father Ted Kilnettle Shrine Tape Dispenser (stephencoyle.net)

Cholesterol treatment can cut levels by 69% after one dose (sciencefocus.com)

Turn: Washington's Spies review (2016) (passionforthepast.blogspot.com)

Welcome to the age of $10/month Lakehouses (tobilg.com)

Reviving Astoria – Windows's Lost Android (trungnt2910.com)

It's Waymo's World. We're All Just Riding in It. 10M Rides Surpassed (wsj.com)

More than half of top mental health TikToks contain misinformation, study (theguardian.com)

Insane web design on Apple dot com (lapcatsoftware.com)

Structured Errors in Go (southcla.ws)

Show HN: I built a Atl-Tab like UI for ChatGPT to jump to questions quickly (chromewebstore.google.com)

MIT Class President Barred from Graduation Ceremony After Pro-Palestine Speech (nytimes.com)

Why Use Structured Errors in Rust Applications? (home.expurple.me)

The Prominent AI Events of May – Full Timeline (nhlocal.github.io)

Trump drops NASA nominee Jared Isaacman, scrapping Elon Musk's pick (theguardian.com)

How to Report Bugs Effectively (1999) (chiark.greenend.org.uk)

Tldx – CLI tool for fast domain name discovery (github.com)

Beyond Attention: Toward Machines with Intrinsic Higher Mental States (arxiv.org)

AI Didn't Kill Stack Overflow (infoworld.com)

Ask HN: How do you improve code for future AI?

Comments (3)