Comment by user43928
4 days ago
If you were to take away my work's $500 budget for Opus 5.5 and replace it with unlimited use of GLM-5.3, it would be a huge downgrade in my opinion.
If your employer is so cheap as to spend a $100 or less per month per engineer, while being large enough not to have access to the subscription plans, a move to self hosted or cheap providers does make sense in my view.
> If you were to take away my work's $500 budget for Opus 5.5 and replace it with unlimited use of GLM-5.3, it would be a huge downgrade in my opinion.
I am perfectly able to do the tasks with own custom harness and unlimited local Qwen 3.8 27B with 150t/s. Harness needs search capability and more fine-tuning for the job.
You need less words to describe the intent with Astra/Opus 5.5, and it automatically usually observes the best practice that you don't need to explicitly tell, but with only a slight more effort you can get the job done. I have even made security audits with Qwen 3.8 27B against projects fully written with Claude Opus and it is a bit mind-boggling, but there have been tens of vulnerabilities, identified by this model alone. This excludes the issues found when guiding more myself.
I’ve been going back and forth between opus and local Qwen 27B and a good number of times opus makes absolutely stupid things compared to Qwen.
For a large number of tasks you definitely don’t need a frontier model but a good harness around it.
If you don’t can’t/want to experiment with local model, have Claude implement the exact same feature with haiku/sonnet/opus and see what the differences are. You might be surprised.
I've used extensively everything between Opus 4.6, GPT-5.5, Fable 5, Opus 5, GPT-5.6 Sol, GPT-6 Astra, 6.1 Sol and Opus 5.5.
To a lesser degree also Sonnet and Luna.
And when I say extensively I mean 500h+ on a side project in addition to my full-time job.
Even the difference between 5.6 Sol, which is miles ahead of local Qwen 3.8 27B, and Opus 5.5 is immediately obvious in the quality of the results.
Having been stuck on 5.6 Sol at my job while Opus 5.5 was available at home, I can say with full confidence that the difference is significant, even if I weren't looking at the output at all.
The way it understands my requests, forms a hypothesis, verifies the result and finishes the change is night and day different. This happens while 5.6 Sol is still flailing around somewhere in the monorepo struggling with even locating the relevant code.
I took a break from commercial plans and was using local qwen/glimmer/gemma and hosted glm/DS4.1-flash for things; opus 5.5 is night and day difference to return to. It makes some mistakes certainly, but its commits are often exactly the change I would have made with far less babysitting. It also shows that fable spark of common sense of “hey…you realize this other thing is broken too right?” without getting lost in the weeds immediately like other models seem do. Still an idiot sometimes though... “Oh no, I’ve discovered the exact same thing we have discussed and you have approved 5 times, and I’ve made notes about…I don’t know what to do, can you tell me to read my own note I’ve just reviewed and ignored?”
1 reply →
I use a mix of GPT/Codex, Claude, and DeepSeek for my workload (in that order of volume).
Each has their own characteristics and honestly you just get used to prompting each a bit differently.
Even better is to use an orchestrator (I switched to Paseo a few weeks back) with an agent "team" (in Paseo, these are just profiles that the orchestrator initiates). Then the orchestrator writes super detailed prompts to each sub-agent on your behalf and I find I get better results this way with less effort. I would describe it as "more autonomy".
In my company they are also hosted GLM 5.3 in their on-prem have us a unlimited tokens. It does a decent job, I would say.
But why do you think it's not good apart from budget?
The thing is the have been told to use the tokens wisely, but no limit is communicated. Still everybody is hesitant. To be fair, the company runs claude on AWS bedrock for data privacy requirements at a different pricelevel. But I guess that is the drawback for german job-security: each and all of them know they basically can't get laid off.
I also attribute a lot of it to be just plain outright denial. I have a feeling a lot of people don't WANT to understand that their coding skills they acquired over years have dropped a lot in worth. It is the "as long as I don't step on a scale, I can't be overweight" kind of denial. You are afraid to see what is behind the curtain, so you simply don't do it.
> But I guess that is the drawback for german job-security: each and all of them know they basically can't get laid off.
Help me understand what exactly is the drawback you’re mentioning. In your previous comment you were suggesting these people spend their own personal money and personal time on projects to benefit their work time. They (quite reasonably, it seems from the limited context) decided that no, they won’t dedicate life and resources from outside work to more work. Fortunately, it seems they have some job security (German, but probably some from the EU as well) which allows them to make whatever choice they believe is healthier to them, without fearing repercussions. What’s the drawback?
The obvious drawback is that according to his description, they aren't getting much done and neither are they producing quality work.
This would check out with the fact that there aren't many software products or platforms made in Germany, compared to say the US or China.
This raises questions about how sustainable this situation is and if it might further increase the gap in income of software engineers in the US compared to Germany.
Even the best benefits do arguably not make up for receiving $2500 net at the end of the month compared to $8k in the US.
And I get what you are saying about work life balance and so on, but if one cannot even be bothered to try a $20 AI subscription at home, there must be about zero passion for the craft.
5 replies →
No you misunderstood my comment. I suggested they spent their private money - 20€ a month for the lowest Max flatrate tier - because to me it is obvious that you absolutely need to develop the skill to work with AI. They only have the "pay-per-token" model on their work accounts and laptops, because that is what large corporations get - no flatrate. That leads them to not work with AI. Most of them just refuse to get that 20€/month subscription to experiment with it on their private laptops. I didn't say private time - company policy is no US cloud services for work related topics so they would have to install it on their private computers. They totally could work on toy projects during their time, as the team is in homeoffice a couple a days per week.
I mean the drawback is none as long as they have job security. Unless you realize you work for a public entity that is paid for by the tax payer.
11 replies →
What exactly is the skill you acquire by using more AI? If the models get smarter, isnt it a waste of time to get good at the current gen?
I'm probably a worse programmer than I was before I started using frontier AI models, in the sense that I probably have a harder time reading, writing, and debugging code. And I doubt that my knowledge of the AI models themselves will have lasting practical value.
But, because of the explorations I've done with them, I know substantially more about input method APIs, UEFI, PCIDs, seL4, WebAssembly, GLSL, Service Workers, CORS, server-sent events, CSS, content-security policies, memory-hard PBKDFs, Android, numerical integration, Verilog, V8 security, Linux process namespaces, hypervisors with VT-x, multitouch user interface design, radio communications theory, power electronics, TypeScript, physical-layer communications protocols, NTSC video signals, TLA+, bounded model checking, Rust, generative testing, theorem provers, FPGAs, Linux system call overhead, Unix mailbox formats, physical modeling synthesis, damped resonance, KaTeX, geology, dishwashing detergent, microbiology, Cyclone, thermodynamics, X-Windows, localStorage, AVX, Excel, and IndexedDB, for example. Most of these are things I knew very little about before, so I may not know much about then now, but it's enormously more than I did.
Aside from being very pleasurable, I have the impression that this knowledge is valuable in designing systems whose implementation I can then elicit from an AI model.
I don't know how long the knowledge will remain instrumentally valuable. Possibly a year from now, we'll all be dead, or slaves to Sam Altman. Or the AIs will be so much better at designing the systems we would have wanted to design, if we were smarter and knew more, that there's no instrumental value in humans doing it. Or, less drastically, maybe a new AI-written software platform will have so completely displaced Android and the browser platform that there's no value in knowing what they do and don't permit. Like knowing the ins and outs of PHIGS, INFORMIX-ONLINE, or CORBA. (Thermodynamics is probably safe, though.)
But it seems like, in the less insane possible futures, a lot of this knowledge will remain empowering for some time.
3 replies →
The skill of getting familiar with AI use, how to use it, when to use, learning when NOT to use it, is the skill that will land you your next job. Or would you in 2030 expect to hire software devs that never touched AI?
1 reply →
You get better at building things - asking the right questions at the right time, knowing what to trust and what to look into.
Not very different from managing a team, just without the human problems.
That is a really silly corporate policy - if the actual amount the company wants each employee to spend is maximum €200/mo, $500/mo or whatever then the company should set that as a limit and people will use it.
Saying "you can use as much as you want but don't use too much or you'll get in trouble" is obviously confusing and is going to result in most people barely using it
Er, for me it would be a massive upgrade. Hopefully my employer follows this lead.
[dead]