And if you need more coverage or other networking gear at a later point, you can just tack on another access point or a switch or something, and it all works together seamlessly through a single interface
I've been running Unifi since ~2014 and it's been rock solid. I forget that they are there most of the time.
The only headache has been setting up a Canon printer -- it doesn't know what fast roaming is and won't prompt for a password if it's enabled. You must disable that first, connect the printer, and then turn it back on.
I have tens of thousands in unifi gear, it is pretty damn good but there are bugs, and unfortunately their support sucks. It is still miles ahead of any other consumer networking hardware.
It's not difficult to avoid the 1000 lines per PR thing: Depending on what kind of thing I am adding, the plan might also receive as instructions to value making as small a code change as possible. It still requires judgement, as on something big, the smallest possible code base is not necessarily the most readable, but this is the kind of thing one can decide with some experience and little work.
I've also managed to use LLMs to cut a lot of manual duplication in code where we typically didn't do enough investment: "Claude, evaluate code duplication in the functional test suite" will have no problem finding things like insufficient helpers, or tests that are testing simpler things as prerequisites, so they can rely on each other. So I am not seeing my codebases growing all that much. There's some risks of functional changes that before would be rejected due to cost which now are not, but I am not all that sure of how much that is controllable without being relatively antagonistic with management.
It doesn't engineer. It writes code. Enthusiastically. Usually without thinking about the bigger picture, the design, the architecture, the trade-offs, etc.
It's up to us to manage that process.
It's why senior engineers are finding LLMs a really useful tool - because we've learned to think about all that other stuff before opening the text editor. Writing the actual code was always the easy (and least valuable) bit.
There's definitely a way to use Claude code that is token conscious.
I've tried throwing unsupervised agentic software factory workflows against the wall, and they burned through my tokens like nobody's business but didn't produce much.
Supervised, human-in-the-loop process on the other hand is much more productive but doesn't consume nearly as much. Maybe that's why everyone's pushing agentic approaches so much.
The current thinking is automated agents is what turns this from an industry in the tens of billions to a multi trillion dollar one. So yes you are right on the money, agents stimulate demand for this thing they've built.
AI is expanding to meet the needs of expanding AI. Why worry about jobs? AI will provide plenty of work. If anything, I worry we'll be working more, not less. All that AI will need someone to vouch for it and to scapegoat when it makes mistakes.
There is always a quantity of lubricant that can get any machine moving. Just add so much that you create an all consuming river of lube and watch your thing sail away.
Yeah. Claude does good work but reviewing it all properly takes quite a bit of time. It got to the point I started having trouble maxing out my weekly allocation.
Dealt with that by going all out and making an agentic parallel code review skill. Basically an infinite TODO list generator. Now I'm definitely getting 100% of the usage I paid for. It really burns tokens like nobody's business, and catches a lot of issues while at it. I've been looping this review/fix process every week. It's dramatically reduced the amount of stuff I need to pay attention to during my human review sessions.
I really don't like how the payment plans work with the providers right now. I feel this pressure to use all my tokens for the week, often just "wasting" them. But also, I want to take advantaged of the subsidized tokens in Claude Code and Codex for as long as I can.
There is this real danger that our thinking, and the things we make, become bloated without constraints.
IMO software has gone to shit since both mobile phones and laptops mostly have massive amounts of compute. We always seem to use it to the limit, just because it's there.
It's the gacha of software development. We've got periodically resetting timers. Prompting is like booster packs: we have a finite number of dice rolls before the timer resets. We might even get a Legendary Ultra Rare pull whenever Claude happens to be feeling extra motivated. Before you even know what's happening, it's hijacked the brain's reward circuits to the point you're waking up at 3 AM because that's when the timer resets. Gotta saturate those timers with pulls and minmax everything in sight.
At least it's doing something productive instead of just sinking money into literal gambling simulators. Mercifully, unlike video games, automation is not "cheating".
That will be true in March 2027, but isn’t necessarily true today. I work in a large organization that is grandfathered into the Enterprise Subscription plan until then. We have thousands of seats using Claude Code.
I’m interested in how this works in practise - I guess you’ve written a skill to do code review, then your Claude.md file tells it to use it after every change as a bg task? So does this work as a background task while Claude is working on the next ‘feature’?
There are many "critics", one for each quality I want reviewed. Correctness, consistency, maintainability, security, testing... Everything I could think of, and I keep adding more.
The scrutinize skill is the entry point. The Opus I'm talking to becomes an agent coordinator. He explores and autodiscovers the project's structure, subdivides it into logical sections.
Then he runs a truly absurd critic x section matrix against the entire project. Literally hundreds of these agents running in parallel, each focusing on one area. Ten minutes of this is enough to exhaust my Max 5x five hour window and put a serious dent in the weekly usage numbers.
It literally takes days to run a full agent sweep. I designed it around the rate limiting. The agents do file system style journaling in order to resume cleanly. They commit all of their findings as they go into an orphan branch in the repository. Further review runs can build on it and avoid searching for known issues.
The way it works in practice is I just run /scrutinize sweep and then go work on something else, or just go do my actual job, live my life, play video games, write an article for my blog or something. Come back five hours later to either resume the process or check the literally hundreds of issues that have been found by all the agents. Then Claude and myself will go in and evaluate and fix all of those issues one by one. Then review again. Then evaluate/fix again. I'm just gonna keep looping this over and over until zero issues are found. For all of my projects.
Going from solo hobbyist programmer to this was pretty insane. I can only imagine what these corporations with infinite money must be doing.
Those critic skills are great. I see a real business opportunity for someone who can bundle everything you describe into a turnkey solution for a programmer like me who doesn't want to take months coming up with their own system and extensive .md files.
I'm currently, very painfully, removing a tiny bit of tech debt at a time from a massively complex project that we inherited from a 3rd-party vendor. Some of the tech debt is AI-related, some because it's a vendor who rarely has to maintain anything they create, some because when we first inherited it we had no grasp on the entire codebase and were just trying to change the plane wheels while flying (we still are).
What I'm doing now is the hardest kind of programming imo. I spend hours/week just meditating on how to chip away at this out-of-control codebase, figuring out how I can surgically remove some leaky abstraction that's spawned 5 cousins w/o disrupting the whole project. I'd be fascinated to see if the latest frontier model with a system like yours can actually help me. But I don't have the time or desire to invest the months of trial and error that I'm sure it took you to get to that point.
I used Claude Code's /insights function. It gets Claude himself to go over your sessions and usage patterns. It'll produce an HTML report that you can view.
In my case Claude saw that code review was my main activity and that I was manually and repeatedly asking claude to "review X, Y, Z..." so he suggested turning it into a skill. So I fired up the superpowers:brainstorming skill and bikeshedded it until I ended up with this heavy duty massively parallel super reviewing super claude. Refined it a bit after a couple weeks of use and the result is what you see in my repository.
You're welcome. Email me when you discover your own path? I could learn a thing or two as well. I'm pretty much a beginner when it comes to this stuff. Subscribed like two months ago.
I would advise against it, depending on the project.
My lone lisp project gets the most love. I spend weeks reading, reviewing, restructuring and rewriting everything. It's the project where I'm concentrating all my efforts. Everything I push to master is absolutely my own work and I do want everyone to read it.
I had no trouble letting Claude take over maintenance of my static site generator and virtual machine orchestration scripts though. I wanted to care but... I didn't. I did glance over the finished product just to ensure it wasn't going to nuke my laptop the second it ran, but that's pretty much the extent of it.
I think it's great. People at a broad scale are getting first hand experience with resource management. It's a fairly cheap way of doing it too (in contrast to: learning this by managing humans) and we can all benefit from the skill transfer.
I find myself observing how my lead manages meetings ... "ah, this is like when I do that with Claude", "this is where he wants to understand what happened, like when I ask Claude" ... it's funny.
At the enterprise level though, its going to be hard to want to use a service in which costs are not predictable, and keeping those costs under control requires employee training.
So your costs scale with the number of users you have.
Thats an op ex that you can explain.
For tokens for developers its maybe closer, cost/outcome wise, to hiring an external consulting company to write your code; money paid scales with work done, no promise of delivery, arbitrary unpredictable external price changes.
Its not quite the same; though, similarly lucrative for consultants.
2 months ago: no limits. 1 month ago we had a leaderboard for whoever had the highest token spend not taking into account what was actually produced. This week: “everyone is using opus too much, just use it for planning.”
i've worked at so many places where the propaganda/marketing and reality on the ground is so disorienting/shocking i don't really expect this to be any different...
since those headlines started ive felt it just encouraged inefficiency. "say as much as you can without saying anything." if you were accomplishing your task the need for more would end, thus there is incentive to never succeed.
You can put a limit on token spend and provide training (and even pre-configured workflows) on how to limit token spend.
Like the other commenter said: cloud spend can also spin out of control if you don't pay attention, yet we've found ways to keep it under control (training, guardrails, limits, transparancy).
The problem that I see is what you do if someone runs out of tokens. It doesn't very well work to say "well I guess you just get fired because you can't work at full speed for the rest of the month".
Personally, this feels like its just trying to push the work of managers in allocating resources onto developers so that they have more work to do and can be blamed if anything goes wrong.
To be fair, the cost of software development has always been fairly unpredictable. What may be different is that the cost used to be roughly proportional to man-hours spent, while now the number of agents running in parallel may be less predictable.
> To be fair, the cost of software development has always been fairly unpredictable.
Yes, but in a "oops this is gonna take another two months to finish" kind of way, not the "oops this is the 12th time this month 8 developers have burned $2K in tokens in a single day and no one really knows how it happened" kind of way.
A belt loaded spinwheel machine gun, where there are some chances the next bullet is a dummy round, or goes in the wrong direction. And everytime you reload a new soldier is in charge of the gun
You don't need that analogy as the normal use of a automatic gun in war is not to kill, it is to suppress - stop the enemy from moving. If you are hit by a gun in automatic mode it is your own stupid fault. When you want to kill someone you switch to one shot or maybe 3 round bursts.
The only important thing to know is they're loud and thus impossible to conceal. Maybe someone out in the open would be hit if one starts firing unexpectedly. However, the vast majority of cases, the purpose of an automatic gun in war is just to put oatness stream of bullets so that everyone on the other side knows it's stupid to show your head and this allows your own people to move in relative safety because they know where your automatic weapon isn't going to be firing and the enemy, of course, isn't going to be able to respond.
The cost per month is 100% known and always has been. What has been variable is the rate of delivery. AI is different and can be substantial in countries with lower wages.
There have also been winners of a slot machine gamba, so the analogy quite holds. I would even argue that there are considerably more slot machine gamba winners than the real world examples of actual LLM work.
There’s actually been a ton of research on how to optimize “slot machines,” at least in a generalized sense. For more reading, check out the literature on multi armed bandits.
Yes, because in video games there is always a chance to win so you can optimize your strategy around that chance. If you have a 1% chance to drop a legendary weapon, the question becomes how do I manufacture 100 chances for a weapon drop in the shortest possible time. With agentic coding there is no such guaranteed chance - in a way it's worse than a slot machine that is guaranteed to pay out eventually. You could spend hundreds of millions of tokens and still not get what you asked for.
> If you have a 1% chance to drop a legendary weapon, the question becomes how do I manufacture 100 chances for a weapon drop in the shortest possible time.
Sidenote but I hope everyone realizes that 100 is kind of arbitrary here and does not mean the total chance to to get something is 100%.
You’re right, the arpg analogy isnt great, it’s too simplistic. I was trying to come up with something heavily stochastic where people are coming up with strategies to get the odds in their favor. Maybe closer to speculating on the real estate market? But even that feels too simplistic compared to LLMs. Even the definition of a win isn’t well defined.
Actually it’s really its own thing, I don’t think the slot machine analogy works too well, you also have fixed odds (and you know they aren’t in your favor), and a binary output
The analogy to slot machine is that you're spending your own resources in hope of a reward. So you're ultimately bound by your resources and your strategy doesn't count for much in the grand scheme of things.
With employees, there's a lot of punishments in place for people to not want to mess up. Loss of wages and reputation, prison time,... Startup do not fail because they have a bug-ridden product, they fail because of the market.
With AI, all bets are off. They're not aligned with your goals and it's very hard to discern when they go off unless you're an expert. And if you are one, at best it's just a slight boost in typing especially with all the works involved in software development.
Odd, I train teams (at large companies) to use harnesses effectively. So some training does exist.
I get the anti/skeptic sentiment. I've been called a lot of horrible things by a vocal contingent when they hear that I help train folks to learn software engineering best practices and then apply AI to that.
> There's definitely a way to use Claude code that is token conscious.
Colleague used Sonnet 4.6 on some pretty normal agentic coding tasks through AWS Bedrock to keep the data in the EU, 100 EUR usage in a single day. In comparison, the Mistral subscription costs about 20 EUR per month and we tested that for similar tasks it was okay, the usage got to around 10% of that monthly limit in a single day. Or Anthropic's own Max (5x) plan where you get way, way more tokens to do with as you please.
I feel like the sweet spot is having a monthly subscription with any of the providers (you're subsidized a bunch), but if you have to pay per tokens, now I'd just look in the direction of what tasks DeepSeek would be okay for, sadly probably not in the situation above. For a startup, though...
On the other hand, this feels a bit hypocritical:
> It was part of an effort to get project managers, designers, and other employees to experiment with coding for the first time, and sources tell me that Claude Code has proved very popular inside Microsoft over the past six months.
They're gonna say that the future is all AI... until they get the bill.
I was a Mistral Le Chat Pro subscriber (the €20/month plan). Yesterday I hit my monthly limit. Switching to PAYG I burned through another €40 in one evening, working on the same project, with the same tasks.
I upgraded my plan last night to Mistral Le Chat Teams. This now costs me €60 per month for two users. Limits have been reset, but I have no idea now if my per seat limit is higher than the Pro plan, or if the limit is shared between the seats, it’s really not clear. I guess I will find out next month. The limits reset on the first of the month and I really hope I don’t hit them in the next seven days.
I use Mistral Vibe CLI and I’ve written and implemented a couple of new skills[1]. Caveman, based on an idea I found online somewhere, this skill removes all extraneous response text, including articles. Makes for some fun reading, but supposedly reduces output tokens significantly. Hash-anchors, this one is based on a concept from Dirac[2], reduces search failures and also includes multi-file dispatch. It will be hard to measure, but Vibe tells me these two should result in roughly a 40% reduction in token burn.
I was trying to get a better sense of the time cost quality matrix of these, so I threw together a quick eval of Sonnet 4.6, Mistral's dev model, and Opus 4.7 (figuring it's what you'd use if you were on Max).
The results for a function implementation and test of levenshtein distance in js are pretty similar but Mistral is 30x cheaper than Opus 4.7 and 4x faster than Sonnet 4.6.
Levenshtein distance is not only a well-understood problem, it's small, self-contained, and extremely well-represented in the training data. The kind of problem where even small/bad models can excel. The golden standard for those tasks is just "use a library" so no wonder the beefy models are expensive: you're chartering a commercial airplane to go grocery shopping.
My personal benchmarks are software engineering tasks (ideally spanning multiple packages in a monorepo) composed of many small decisions that, compounded, make or break the implementation and long-term maintainability.
There's where even frontier models struggle, which makes comparisons meaningful.
It’s making guesses not decisions, framing as decisions will lead you astray to wasted time and tokens.
It’s vaguely productive to tell them a ton of relevant info upfront attempting to minimise their need for load bearing guesses. I say vaguely because obedience is generally only around the level where it's good enough to lull you into a false sense of security, not to actually be obedient.
It’s a bit more productive to use the various loop mechanisms (hooks, /goal etc) to evaluate each end of turn against guard rails and reject with clear instruction on whats unacceptable. Obviously if you only do this without the front load of info then you’re likely to spend more tokens to reach a satisfactory end of iteration.
While you are correct that something like Antigravity 2 + Opus 4.6 can handle large scale software engineering tasks, I would argue that it is usually (but not always) better "coding agent hygiene" to work on smaller code modules and as the human in the loop be a partner, not someone who prompts and then disengages.
Breaking code up into composable chunks has worked well for me over 50+ years as a professional software developer, and I can't get away from the idea that it is still usually the way to go using agentic coding tools.
The one detail I did forget to mention is that if anyone goes with the Mistral subscription (instead of paying per-token), then the Mistral Vibe tool gives you their Medium 3.5 model by default, with a 200k token context. It will probably be enough for plenty of tasks, though there's also a noticeable difference between that and up to 1M.
My experience as well... I've only hit Antrhopic's 5hr threshold a few times, and two of them was within a half hour of the window. Also, all three times I'd already accomplished a LOT.
I tend to work with the agent, and observe what's going on as well as review/test and work through results/changes. I spend a lot more time planning tasks/features than the execution, even using the agent as part of planning and pre-documentation. It works really well. I don't think people burning through the 5hr allotment in under an hour are actually reviewing/QC/QA the results of what they're doing in any meaningful way, and likely producing as much garbage as good (slop).
I'm really curious as to HOW the MS employees were using the agents as much as what they were doing.
I suspect subscription limits are quite a bit higher than the equivalent tokens their dollar cost could purchase. I similarly feel like I can get a lot done with a $20/mo Claude Pro subscriptions, but also can easily spend $10-20/day at API pricing with similar usage.
Personally I prefer the API pricing because I feel like I'm not going to get rug pulled on my work. When it comes to personal stuff, I use the shit out of my sub, but it's not making me money.
I’ve made the same argument On Here. Paying the full price (should!) make you consider you usage, pick the right model, delegate to cheaper/local providers, …. It makes you use the models the way they’re going to be used after the subsidy ends.
Depends on what you're optimizing for. I'd hope that "after the subsidy ends", the "cheaper/local providers" will be at the level of at least current SOTA models. If not, then there's hardly a point using them anyway; if yes, then by sticking to subscription workflow you'll be learning the very workflow you'll be using "after subsidy ends".
Either way, I don't see much point of intentional austerity in times of extreme growth. There will be time for austerity once the growth ends.
Because with Max subscriptions, you have to use the Claude Agent SDK, which is basically running Claude Code underneath. You don't get to use the chat/Messages APIs with personal subscriptions, for that you need the API pricing.
Terms of service prohibit subscriptions for employees of companies bigger than X people. I suppose they could all sign up as individuals and try to get away with it but presumably that would look pretty obvious with a tiny bit of analytics.
98.6% cache hits doesn't distinguish an efficient workflow from an overly chatty linear agent repeatedly reusing the same context. Plus, it says nothing directly that the process has good useful progress per token.
I doubt it, the difference between someone slightly inefficient and someone extremely efficient isn't big enough to matter compared to how much they cost in salary.
You pay for cache hits on every turn and even with the newest architectures longer context is slower/more energy intensive. Constructing concise turns that reuse prefix and stop when the new context is no longer useful help, as does pushing generation down into cheaper models while using stronger models for verification.
> 2024-01-05 status update: With my 2023-12-04 layoff from Spotify I lost the internal data-access required for ongoing updates to many parts of this site. Most of this, as a result, is now a static snapshot of what, for now, will be the final state from the site's 10-year history and evolution..
what a shame. I didn't realize the author worked for Spotify. Guess it makes sense. Spotify should've acquired it from the author or made a deal with him to keep it live since all the links lead to Spotify anyway.
You’re exactly right..that’s how I end up learning tech stuff. It’s not working very well with modelling though. I’ve set my sights too high, the projects are too involved
I found zeroing in on a specific package and learning the specifics of the parametric modeling AND a particular end goal to be useful. Doubly so if the YouTuber is making lots of videos on other topics that might be helpful. I learned on an olllld Fusion360 tutorial series from NYC CNC, but my more recent FreeCAD learning has been from MangoJelly: https://www.youtube.com/@MangoJellySolutions
TinkerCAD is easy to get started with and is a fine starting point. If you want to go farther than TinkerCAD can reasonably take you, I’ve done a lot of modeling in onshape (after using Fusion360 for a few years). Parametric CAD systems (Fusion, onshape, solidworks, freeCAD) feel like a nice intersection of design and programming to me.
My current workflow is onshape for all modeling (because it has excellent multi-player concurrently editing support, relevant to FRC robotics), then Fusion for CAM¹ if it’s going to CNC equipment. Onshape added support for CAM in the last 8 months, but I haven’t switched yet.
¹ - Computer Aided Manufacturing-turn a shape into a series of instructions for the CNC equipment. Roughly:
I originally bought the touchpad for my UHK. But, much to my surprise, I have gravitated towards the keyboard's built-in mouse layer over time! Now I scarcely plug in the touchpad (or even key cluster) modules at all.
As a sidenote, I love my UHK. Just a joy to use, and it's so easy to customize. I don't have any experience with competitors like the ZSA Voyager, but the UHK's configuration software and macro language do make it quite pleasant to bend to your will. For instance, I do some funky stuff with macros and lighting here: https://www.cgl.sh/blog/posts/wnl.html
I had a UHK for a few months before refunding it because the shielding wasn't good enough to stop it becoming unreliable when my mobile phone was within about 30cm of my keyboard. I contacted support and the solution was to move the phone away from the keyboard, which is kind of irritating for such an expensive piece of kit.
But, just wanted to share that I was similarly surprised to land on mouse keys as a preference. I tried most of the UHK modules which were also pretty good and have since tried various other trackballs and pads, but since trying UHK mouse keys, they're what I keep coming back to most, even since switching to new keyboards.
One issue I have with mouse keys is fear of using them in front of others though: every so often, if I need to click something particularly small and don't have a keyboard shortcut memorised (vscode panel resizing is one) it can sometimes take me a fair few embarrassing seconds drawing small squares around my target before I resort to actual mouse hardware.
For the amount of time and thought and effort people have put into alternative mice, I feel mouse keys are massively overlooked and probably have a lot of room for software/firmware innovation without hardware costs.
You bring up a good point -- I have the same issue with mouse keys. I wonder how the track point gets around this. Is the tracepoint "progressive" in that it allows various speeds depends on deflection from center?
That's a good question and make me think. I always thought the trackpoint nubs were binary too, basically just a stick in the middle of up/down/left/right mousekey buttons, bit it turns out they're not!
For original trackpoints, it's basically a stick in the middle of an up/down/left/right resistive strain gauge.
For the ploopy beans here, they use hall effect sensors instead of resistive strain to get a bit more movement.
As soon as you have non-binary up/down/left/right values, the mouse direction and speed can be interpolated to so many values that mousekey accidental squares become impossible.
It was however stuck in customs for quite a while, but I guess that's what I get for ordering custom electronics from Israel right after those pagers blew up.
The UHK also supports mouse keys. I own both the 60v2 and the newer UHK 80 (I was a beta tester), these keyboards are my final destination. The only reason I would consider a change is if they released an ortholinear layout.
Yes, of course. I’ve accidentally fallen down the split keyboard rabbit hole and now I can’t get out. UHK seems like it’s got everything that I might ever need but I’m going to stick with my Aliexpress Corne for a bit longer before I take the plunge.
I went the opposite way: I started with UHK, then went for a ZSA moonlander, but settled on a kbdcraft Israfel, which is a relatively cheap, split ortholinear.
I felt most of the extra functionality and polish that I guess makes up the massive costs of UHK and ZSA wasn't actually necessary. It was cool and fun and useful to try a bunch of different stuff, but then over time, I wanted things to be simple and small which UHK and ZSA Moonlander aren't (ZSA voyager wasn't at the time).
All I'm saying is if you've got comfortable with a cheap Corne, I think you might feel underwhelmed if you spend a lot on something a lot fancier.
I have a tented split scissor switch that is quite wide, and it still annoys my shoulder with the inward pivot. Plus I seem to do reasonably well two fingering a smart phone so, well, what is the point? We can put a SOC in an interface with reasonable speed these days. I would love to see a radical departure. A glove interface surely could be better?
No problem! The only helping hands I'd recommend are https://omnifixo.com/. Pricy, but everything else I've tried has too much play and is useless to me. Omnifixo is just fantastic.
reply