"genuinely" is an AI tell now as well as doing things in physical world that don't make sense like walking to the car wash to wash your car if it's close, or maybe not using USB ports in the way they were designed...
This means that third-party tools built on the Agent SDK like Conductor and OpenClaw work with your Claude plan, but will draw from your credit the same way your own scripts do.
Monthly credit amounts vary by plan:
Pro: $20
Max 5x: $100
Max 20x: $200
Team Standard: $20/seat
Team Premium: $100/seat
Enterprise: Varies by seat type
After you claim the credit, it resets with each billing cycle. Credits do not rollover.
fwiw I've had my site on front page hn a couple of times. It is a completely unoptimized hetzner server running nginx and serving HTML.
computers are _fast_ these days, you're more likely to have an outage from cloudflare than by just skipping it IMO (for basic personal sites, like yours seems to be)
Not really, even "legit" marketing providers have massive automation rigs to warm email addresses, make them behave naturally and email each other in rings for a bit before using them for cold outreach.
So they'd just do this to farm invites if they needed
This kinda thing makes me sad that keybase sold out to zoom and wonder if it can be resurrected. It was such a simple web of trust that went viral enough that I still occasionally see it on HN or Twitter profiles even though it's been long dead.
There are maybe 20 or so online handles I know, some of whom I've met in person, who I deeply trust. To the extent that I fully trust anyone they vouch for too.
Even with just one degree, that's a large enough international semi anonymous online community that can provide value to each other through online text based communication. Doesn't need iris scans or credit card checks, just "patio11 on hn Twitter and whatever his domain is is one of the good uns" and a network effect from there.
Already seeing some form of this reputation staking in eg Pi PRs, everyone is treated as clanker slop by default but the entry bar remains quite low to prove and build reputation.
I don't think online communities will stay the same in the face of AI but I do think whatever comes next will strongly rhyme
Dresden is truly blessed with cinemas and has four European Network cinemas. Three of those have assigned seating, though none do price discrimination based on where you sit. Culturally the assigned seating isn’t taken very seriously in those four cinemas, though, to the point where staff in one cinema sometimes tells visitors that they can sit somewhere else if they want to. In practice we still try to get seats where we want to sit and stick to them (front/middle, away from other people), though if people come in and sit right behind us we might change rows.
With new ticketing systems and online booking being introduced I think there has been a shift towards assigned seating. I remember the first time I was in a Dresden European Network cinema (Schauburg in 2015, that’s the oldest cinema in Dresden, 1927) and there either being no assigned seating or a seat printed on the ticket that no one cared about. We also weren’t asked where we wanted to sit. That has changed with a new ticketing system and now we are always asked about where we want to sit.
I think these ticketing systems come with assigned seating and that’s also a factor in assigned seating being introduced.
Notably, the one cinema that doesn’t have assigned seating also doesn’t offer online booking or reservations at all.
The four big multiplex cinemas in the city have assigned seating and do price discrimination based on where you sit – so it’s taken somewhat more seriously there.
So, yeah, my guess would be that the role online ticketing and the respective software/service/devices those cinemas use for that do all play a role in what role assigned seating plays and those can also trigger a cultural shift from sit where you want to assigned seating. (I have vivid childhood memories of my hometown long before online booking with price discrimination sections but no assigned seating in cinemas.)
I've had completely the opposite experience at Ritza, my dev rel agency, for the last 5 years.
My favourite is to work directly with owners, cofounders, etc. They know what they want, and they have the authority to approve stuff - it's really nice just focusing on the 'real' work instead of the admin and to know that the goals they set are likely valuable and not just someone trying to check a box somewhere.
From my non objective, not looking but I try to stay as informed as possible across South Africa, Europe, US perspective and regularly talk to people on both sides and ask them directly
- it's not as bad as it was in the last several months
- it's still very hard to get noticed, get interviews, etc there's so much noise on both sides that personal references are much more important than front door applications. This was always the case but much more now
- there were previously a lot of jobs for low agency people who were good at doing what they were told and meeting specs, AI is taking these as if you are willing to spend hours per week writing specs and checking results then tokens are better bang for buck than freelance devs now
- approximately all the demand now is for directly AI related plays and even people who get them don't feel secure because the whole industry feels so unstable and bubbly, but there's no money in anything not AI now
> it's not as bad as it was in the last several months
I think this needs to be caveated by the fact that the job market is seasonal. The beginning of the year is usually the best part for jobs since departments have shiny new budgets, with a boost again around March-April as people quit after getting their bonuses triggering of another set of churn.
This matches my experience as well as a sr. Ruby/Python engineer (NE USA based)
The market was pretty barren in late 2025, but wasn't too terrible in Jan 2026 and I progressed nicely through two interview processes. Got an offer from one. Made it to the final round with the other but timeframes did quite not align.
shiny new budgets
I was curious about this aspect. Lot of companies have fiscal years (and, presumably, budgets) that do not start at the same time as the calendar year.
This one's trickier — it doesn't have the same crisp metaphor-driven structure as the Lynch piece. The voice here is more conversational, a bit more meandering, and self-aware about its own contradictions ("how do we reconcile...").
A few guesses:
Henrik Karlsson — he writes thoughtfully about writing and AI, but his style is usually more lyrical and essayistic than this. Probably not.
Simon Willison — possible. He writes a lot about LLMs, is balanced about their usefulness, and has a casual blog voice. But he tends to be more technical and specific.
Scott Alexander — no, too short and the rhythm is off.
Paul Graham — the directness fits, but PG doesn't really write about LLM slop in this register.
My best guess is someone in the rationalist/tech-blogger AI-commentary space — possibly Gergely Orosz, Nat Eliason, or Dan Shipper at Every. The "I still type everything myself" framing followed by a defense of AI writing quality has a Dan Shipper-ish flavor to it.
But honestly, I'm less confident here. Want me to search for a distinctive phrase to identify it?
---
I'd say all of those people have significantly different styles so I think Opus is relying heavily on topic and skewing towards very prolific writers in its guesses
> I'd say all of those people have significantly different styles so I think Opus is relying heavily on topic and skewing towards very prolific writers in its guesses
In other words, that there's a bit of Akinator to how Claude is doing so well at identifying famous or somewhat-famous online writers. And of course it's not surprising that a machine-learning system will take every opportunity left open to it to "cheat". OTOH there are things like the "Large-scale online deanonymization with LLMs" paper https://arxiv.org/abs/2602.16800 which seem to show that current LLMs really can deanonymise many or most ordinary posters based on prose style, though I'm not able to evaluate those claims myself. Do we know whether the LLM providers have actively tried to steer their (easily-accessible) systems away from being able or being willing to do mass deanonymisation?
reply