frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Open in hackernews

Harness Engineering

https://github.com/lopopolo/harness-engineering
56•handfuloflight•7h ago

Comments

lopopolo•6h ago
Hi folks, author here and also author of the seminal OpenAI blog on this topic. Let me know how I can help you all let it rip.
jpitz•4h ago
This is interesting. I'd already had a conversation with my harness ( pi ) about incorporating continuous improvement. This is a great deal better than what I came up with.
lopopolo•4h ago
Glad to hear it! Good luck, have fun. The agents tend to do a pretty good job incorporating these ideas. This was an unexpected thing we learned when publishing the initial harness engineering post.
mips_avatar•4h ago
One challenge/opportunity I've had is harnessing really wide running cheap agents. Any thoughts on how to move really cheap agents beyond basic summarization so we can go broader than the pricing of frontier llms allows?
lopopolo•4h ago
Your “really cheap” agents can’t be so cheap that they do not have good tool calling skills. But! Using bigger models to put guardrails in place as static verifiers allows lower complexity changes to “self steer” as tests fail, which means coming down on the cost curve is more effective.
mcapodici•3h ago
I was experimenting and found deepseek-v4-flash and found it cheap (way cheap compared to sota models) and perfectly good at tool calling. I did a post on it https://martincapodici.com/2026/07/18/weekly-ai-learnings-3/

So $1.5 for 40m tokens I guess would cost much more with sota (but would need less tokens perhaps).

hankbond•4h ago
How do you view harness engineering as an organic development that emerges from its use within a specific domain? Basically the meta-loop that allows an agent to tailor its harness to improve outcomes based on performance feedback. I use Pi a lot and I'm very interested in "self-assembling software".

One concrete example might be maintaining a conventions document per-project that covers how to name things semantically from a list of nouns and verbs. The idea is that LLMs are often not very globally aware, but it's important to maintain coherence across a code base in order for it to scale (in size and over time). Sometimes an LLM might call the same concept a Materialization, sometimes a Projection, and its not useful if its using two terms interchangeably without purpose.

Basically, how are you maintaining coherence when there isn't a human steering the code beyond providing requirements and validation directives?

I see you have relevant context in the repo like https://github.com/lopopolo/harness-engineering/tree/trunk/d... but I'm curious what exists beyond context. Do you use any tooling to steer this type of thing more consistently?

lopopolo•3h ago
Your example is super amenable to vibing some tests. As an example, I’ve been able to ban `number` from representing a duration by walking the AST in a linter to fail if var or param names that look like the end in millis or ms or sec appear. This is largely good enough. If you see that “drifting” behavior appear more than once, you have enough to stop and force the agent to write some static verifiers that reject all but the option you want. For a closer example, we did this with zod schemes and their corresponding inferred types to be universally ZPascalCase and PascalCase instead of camelCaseSchema and CamelCase
lopopolo•3h ago
And to address your broader question, yes this is a form of RSI and to me a vastly superior approach to fine tuning since it allows adopting new model releases without throwing anything away while still having the same effect on improving adherence to local acceptance criteria.
slopinthebag•3h ago
What motivated you to quote your own quote (??) in your readme claiming to boost productivity by 100x, and where did you derive that number from?

"When a quote sounds profound enough, reality usually nods out of politeness, without echoes is just a sentence wearing pajamas." - slopinthebag

kubb•3h ago
I would also love to know. Sounds dubious to say the least.
esafak•3h ago
He's reached the singularity. Just kept pointing the agent at its own output, and leveled up...
burnto•11m ago
Don’t worry, the 100x boosted harness wrote it
hahahaa•3h ago
Thanks. I am guessing you have to try stuff and build tacit experience. No other way, just get stuck in and try stuff, then try and learn bits from others?
lopopolo•3h ago
Basically yea. It is the only way to learn how to outrun your priors on what “high ambition” looks like. The labor that goes into implementation is an uncapped resource now.
lopopolo•3h ago
The models are very good now so the feedback cycle on these meta adjustments is much tighter. Yesterday I was able to one shot a Liquid Glass, HIG-compliant and localized DICOM image viewer (frame by frame and looping video) with Apple Intelligence for de-jargoning the series details. Took 30 minutes. But the app had 60% CPU because it was not caching the decoded JPEGs. I can do a point in time fix for that of course, but the more interesting thing is why that misaligned code was permitted to be generated in a “done” artifact in the first place. What other misaligned code from a perf perspective might there be? And how do I intervene into the system that produced this software to make these misalignments statically not meet acceptance criteria?
bagels•5h ago
The mother of all prompt injections.

Do others at OpenAI use this?

ricardobeat•3h ago
What percentage of this was written with AI?

> A command can feel ambient without making its credential model-visible.

I'd be wary of following guidelines on how to build human-computer interaction systems that weren't written with full human oversight. This kind of wording makes me wonder if these are actual recommendations or just the AI pattern-matching and hallucinating something that looks coherent.

lopopolo•3h ago
You can read the receipts in the linked source material yourself. I’ve also read and signed off on every word in this anthology. Try it before you knock it?

But yes, 100% of the content of this repository was written by GPT-5.6 Sol Ultra.

chrisweekly•2h ago
"weary" means "tired", you meant "wary" (guarded against or cautious and skeptical)
segmondy•2h ago
I don't even know how to feel about this after glancing through the repo.
hankbond•2h ago
Yes that's a great example! I think this type of "structural" harness engineering is what people need to be focusing on at this point in the journey. I think in almost all cases "agentic engineering" really is just "good software engineering practices enforced adversarially". Without guards, even the best models still behave like that Patrick Not My Wallet meme.

Transcribe.cpp

https://workshop.cjpais.com/projects/transcribe-cpp
364•sebjones•6h ago•62 comments

Speech Recognition and TTS in less than 500kb

https://github.com/moonshine-ai/moonshine/tree/main/micro
398•petewarden•4d ago•43 comments

FDA approves new kind of cholesterol pill

https://www.fda.gov/news-events/press-announcements/fda-approves-first-oral-pcsk9-inhibitor-lower...
67•mgh2•4h ago•39 comments

Better and Cheaper Than IPTV

https://github.com/stupside/castor
136•xonery•5h ago•36 comments

Mathematicians still don't know the fastest way to multiply numbers

https://www.scientificamerican.com/article/mathematicians-still-dont-know-the-fastest-way-to-mult...
63•beardyw•5d ago•41 comments

Scrying the AMD GFX1250 LLVM Tea Leaves

https://chipsandcheese.com/p/scrying-the-amd-gfx1250-llvm-tea
8•mfiguiere•1h ago•0 comments

If You Build It, They Will Come

https://www.benlandautaylor.com/p/if-you-build-it-they-will-come
360•barry-cotter•15h ago•127 comments

Making Software: How to make a font

https://www.makingsoftware.com/chapters/how-to-make-a-font
22•Garbage•4d ago•2 comments

GPT-5.6 used a prompt to close a 30-year gap in convex optimization

https://old.reddit.com/r/math/comments/1uxj3cy/after_openais_cdc_proof_announcement_gpt56_used_a/
543•mbustamanter•17h ago•347 comments

Classic Amiga titles, free to download

https://amigafreeware.downer.tech/
98•doener•8h ago•13 comments

A Visual Catalog of Retro Macintosh Software

https://www.marciot.com/mac68k-visual-catalog/
27•zdw•1w ago•5 comments

Hardcore IndieWeb: Run your own website 100% independently for only $0.01/day

https://www.neatnik.net/hardcore-indieweb
132•cdrnsf•9h ago•85 comments

I'm Making Strandfall, a Solarpunk Orienteering Larp

https://mssv.net/2026/04/29/im-making-strandfall-a-solarpunk-orienteering-larp/
133•surprisetalk•5d ago•18 comments

Supabase just announced searchable encryption

https://supabase.com/blog/searchable-field-level-encryption-with-cipherstash
7•dandraper•3d ago•0 comments

Mayor Mamdani Says Landlords Can't Use AI Images to Advertise

https://petapixel.com/2026/07/16/mayor-mamdani-says-landlords-cant-secretly-use-ai-images-to-adve...
407•gnabgib•8h ago•182 comments

AI Mania Is Eviscerating Global Decision-Making

https://ludic.mataroa.blog/blog/ai-mania-is-eviscerating-global-decision-making/#fnref:3
148•subset•5h ago•59 comments

Is this the end of the once-mighty GoPro?

https://amateurphotographer.com/latest/photo-news/going-going-gone-is-this-the-end-of-the-once-mi...
213•aanet•4d ago•458 comments

Deepsec

https://github.com/vercel-labs/deepsec
17•handfuloflight•5h ago•0 comments

Fable 5 vs. GPT-5.6 Sol on an NP-Hard Problem: Does /goal help?

https://charlesazam.com/blog/fable-5-gpt-5-6-sol-goal/
233•couAUIA•19h ago•112 comments

I built a browser-based P2P file transfer tool using WebRTC

https://airdows.com/
12•SamOkampo•4h ago•10 comments

Setting up your spare Mac for Claude Code to control, a step-by-step guide

https://ykdojo.github.io/claude-controls-mac/
215•ykev•14h ago•142 comments

Parsoid

https://www.mediawiki.org/wiki/Parsoid
16•debo_•1w ago•0 comments

Gleam Is Now on Tangled

https://tangled.org/gleam.run/gleam
227•nerdypepper•15h ago•142 comments

Elixir-lang.org has a new design

https://elixir-lang.org/
213•bbg2401•15h ago•122 comments

Co-evolution of self-replication and function in a digital primordial soup

https://arxiv.org/abs/2607.09211
56•vicgalle_•9h ago•7 comments

LG monitors silently install software through Windows Update without consent

https://videocardz.com/newz/lg-monitors-silently-install-software-through-windows-update-without-...
1089•baranul•20h ago•555 comments

Developing an Intuitive Sense of Scale

https://magworld.pw
20•vismit2000•4d ago•4 comments

Codex Resets

https://codex-resets.com/
130•denysvitali•7h ago•97 comments

Show HN: Q3Edit – Edit and play Quake 3 maps in the browser

https://q3edit.com
81•drdator•15h ago•21 comments

A Second-Grade Teacher Revived a Beloved Video Game

https://www.nytimes.com/2026/07/13/style/backyard-baseball-video-game-teacher.html
86•danso•5d ago•28 comments