HNHacker News
TopNewBestAskShowJobs

frangonf

481 karma · joined March 11, 2026

10⁻¹X Engineer frangonf.com
submissionscomments
frangonf··on Webster’s Dictionary: The pleasure of finding things out (2014)
Found this: https://github.com/websterParser/WebsterParser

With websters-1913.dictionary.zip ready to unzip and save into the dictionaries folder and use with the Dictionary app.

frangonf··on Fancy Curl Sites
hey that's pretty curl
frangonf··on Smart people don't think others are stupid
I read and love all from Derek. I like that he always makes you think through, even when starting from simple and stupid ideas.

Personally I tend to dismiss myself as stupid, because many times I detect I'm being misguided by lazyness and not wanting to think things through, which obviously makes me feel so smart that I could spend all day trapped in this stupid paradox.

frangonf··on Ask HN: Why do people say "I built it" when we should say "Claude built it"?
I have some personal projects started with AI from day 0, which I use and come back from time to time to update with new ideas, features, product direction, explicitly choosing deps/tooling, etc. I will never say that I wrote it, but make or build doesn't seem that far fetched. Maybe I just generated it? Incantated? Reverse-entropied it?
frangonf··on Ask HN: What is the current trajectory of a senior engineer?
I'm no senior but I think that if software (and rest of disciplines possible soon to be hit as hard) can still be "wrong", if you can still mess it up even with machine assistance, people doing software still have a place. This doesn't mean that we need to achieve absolute perfection to be deprecated, maybe we just become accepting of machine generated good enough artifacts. I'm betting on humanity infinite nonconformism instincts so I can have a job some more years!
frangonf··on OpenAI engineer arrested in bugging of executive conference room
Does detective bench cancel out felony bench?
frangonf··on Redesigning source control at Anthropic
Good point, I wasn't clear enough about my thoughts here. I tried to say that in a multiple commits/s world the review might not be direct line/hunk review of the code but a more "indirect", abstracted or higher level human oversight, which is how I understand an intent review is.

I'm not saying that Anthropic here close their eyes and tell Claude to ship and that in the future there will be no code review at all, but that if we have 30K merges/day, human review seems to need to be done at some higher level, likely AI assisted like it seems to be done with their Antique code review tool mentioned in the post.

frangonf··on Redesigning source control at Anthropic
Excuse me for throwing up this stupid idea, but if code changes per unit of time keep growing at this rate and are not slowing down with infinite merging demand, it seems that this repo will be in a constant state of continuous change and we constantly roll forward with automated code and obviously no direct human supervision, then why even have source control? Why not just let software be fluid and, like Heraclitus river, never step twice in the same repo.
frangonf··on Gemini 4 Argon
In agy 1.3.4

> Gemini 4 Argon (Medium)

  Unavailable · Gemini 4 Argon is currently available to Ultra plan subscribers.
frangonf··on Character.ai has initiated voluntary Chapter 11 bankruptcy
Character.ai (business model Uber for AI psychosis) takes near 3 billion bag and files for bankruptcy in a move to getting up with the best practices of the big boys of entertainment, as ex character ai ceo moves to Disney (business model: Uber for tax-deductible cultural vandalism and EULA-enforced manslaughter).
frangonf··on Sorry, I'm in a meeting
This has real demand, I ended up in the Gitlab secondary youtube channel one day, ordering by most views, and I found a video [0] of a seemingly normal uninteresting meeting with millions of views. Turns out people were using this as an excuse for looking like they were busy. They disabled comments now but I saved some of them because I thought it was hilarious:

I play this when I need my family to leave me alone - so helpful 4.5K I play this when I pretend to be busy. This meeting is literally about anything you want it to be about lol 1.5K I love that we all arrived at this video for the same reason 2.6K Every now and then I'll pause it at a good point in between them talking to make my own voice heard contributing to the "meeting" lol. Gets the full effect in! And the beautiful thing is their conversation just flows with it like they heard me lol. 562

[0] https://www.youtube.com/watch?v=lBVtvOpU80Q

frangonf··on Ask HN: What's a good title for a software engineer now?
I get it from the linkedin and marketing side (I might need to face this marketing needs too), but why do we think is software engineer/developer not representative anymore? The job from an external observer is basically the same with or without clanker usage, you use some HMI to interact with a computer and shove around logic an data to develop digital running artifacts/products. Do they (the people who are in the market for human software generators) even care? If its a purely marketing thing I guess throwing an AI in there is enough eg: AI Native Software/Product/Platform/YourFavouriteDomain Engineer.
frangonf··on Dots: Always-on agents
So the industry is pushing heavily into openclawing their products, "cutemorphizing" the clanker shape and is slackifying the UX so that we can have the familiar UI/UX for the general public and turn the tools more proactive without leaving them too lost.

I think it's a great approach for enterprise since interacting with the machines as a babysitted pet disposable entity is the meta today with human workers. I'm excited to start my new role next month as tamagotchi engineer.

frangonf··on Nokia Design Archive (2025)
> Strategy2 : Maslow’s_Media Scenario’s Media for a purpose. ‘Need to Know’ Media. Media not only for Entertainment. Foreground and Background – I am in control when I can see and access simultaneous media. Routine and Rituals – I have sensible devices that respond to my daily rhythms. Forecasting and Broadcasting – I can ‘anticipate the future’, future needs and responsibilities.

> Strategy 3: Media_Passions Scenario’s Behaviour Passion Experience Product categories and communication strategies defined by media passions Behavioral passions: I’m Social and Share – Focus is Sociability I Consume - Immersive consumption I Organize – Media management I Collect and Mix – Low threshold creativity

> Trans_Life is a project sponsored by the Nokia Multimedia Business Group, to investigate the convergence and integration of multi media products, and develop scenarios visualizing new use patterns and new product propositions for the near future

https://repo.aalto.fi/uncategorized/IO_3a24c0ae-8f52-4cf4-a7...

frangonf··on AI coding has made CI a bottleneck, so we reworked ours to keep up
My biggest takeaway (as a one man band 20 leagues below this spending time optimizing Gitlab free plan and self hosted runners): they did not worry about any of this until 100M ARR and more 1B valuation[0].

[0] https://linear.app/now/sharing-growth-with-the-people-buildi...

frangonf··on AX – Google’s Open Agentic Orchestrator
Since hearing the word orchestrator in ai context it was clear that ClanKernetes was coming.
frangonf··on Jev-Leftpad
Make sure to vendor this package if you want reliable operation.

    it('trusts Jev when it chooses the wrong amount', async () => {
    answer('space_2');
    await expect(leftPad('jev', 8)).resolves.toBe('  jev');
    });
frangonf··on How OpenAI Used Its Own LLMs to Design Its Jalapeño Chip
As a former EE, attention was all I needed to not get zapped.
frangonf··on BeanShell3 in Development
I will remember today with a smile in face as the day I met Beany[0].

[0] https://beanshell.github.io/beany.html

frangonf··on Tell HN: OpenAI keeps re-enabling the 'allow training' setting
There's also this setting here which I have set to 'Do not train on my content' even though I trust it the same as the 'allow training' one...

https://privacy.openai.com/policies/en/?modal=take-control

frangonf··on Show HN: Copperhead – Cursor for circuit boards
I've imagined bugs releasing magic smoke faster than usual.
frangonf··on Muse – Meta’s personal AI agent
Openclaw creator Peter Steinberg was in talks with Zuck at the beginning of the year before choosing to work for Openai.

I think Meta limiting factor for this product was lack of agentic capable models which they fixed now with aggressive RLHF focused data gathering.

frangonf··on Kimi K3 (2.8T) at 1 token/s on a MacBook Pro, streamed from four SSDs
Around 20s saved by eating on those \n\n.
frangonf··on Ask HN: How do you guys keep up with the most cost-effective AI models?
For evaluating subs you have to put the money and try by yourself to get your own measurement, since the subs quotas and new models are always changing and overall info is unreliable.

20$ of Codex/Google/Chinese gives you some fair amount of usage to test them, and Opencode Go for 10$ lets you try a good amount of models with good quota. I don't use Openrouter because it gets more expensive than the subs, but for testing, swapping and being completely independent, a proxy service is the best solution.

About benchmarks, I usually agree with DeepSWE. Looking at usage rankings of models in Openrouter is also a good signal.

For changing models locally I just use pi (used also opencode in the past) or if the sub does not allow login in external harnesses I just use whatever cli they have, last year there were differences but today they are all good enough for my needs and have basically the same features.

frangonf··on Ask HN: AI for Home Lab Infra?
I would start by using AI to port the tangled mess into as much a tidy IaC as possible, with git backed declarative config tools, to prepare the safe ground where agents have context, can get loose and easily review, lint/check or test in throwaway VM before applying it.

A fleet of Openclaws (or equivalent) with control of machines for 100% hands off management seems like a russian roulette to me as of today, but a 90% handoff to AI for reviewing current config of X in our system, reading docs and configuring a deterministic tool is achievable and somewhat safe as you can easily review and trace what has been done.

frangonf··on Launch HN: OneCLI (YC S26) – OSS sandboxed agent harness for teams
I had the project already starred, I lurk around the sandbox space from time to time, and the project it's a few months old and was already discoverable and in a usable state before today.
frangonf··on Everything you do is being recorded
Everything you think will be recorded via neural decoding from your meta smart glasses and apple neural headsets in the near future.
frangonf··on Why is everyone trying to build a solid-state battery?
Autonomous kamikaze drones don't need spools.
frangonf··on Codeberg Divides
I am a heavy viber that uses a lot of resources in github now after LLMs compared with pre-LLMs for my personal projects.

I find great that there are places that are still human first and sensible about community and free software instead of just providing technical infra. Gitlab and Gitea are already open Github alternatives: open source core projects turned open sourcish business with free as in beer plans that get monetized with user data, loss leadering, business offerings and other sassy business models. Codeberg is deciding to not be that. Other projects can decide different and its users should evaluate the tradeoffs and choose.

frangonf··on Hetzner is working on LLM Inference
I read that data centers in Frankfurt use 40% of city energy, and seems that some fossil fuel based workarounds are being pushed. I hope we do better than the US in the energy and data center gold rush.
← PreviousPage 2 of 4Next →