Ask HN: Corporate Disconnect Between "Tokenmaxxing" and Token Optimization
11 comments
What made accountability tractable for me was treating agent output as untrusted input — the invariants I own (cost caps, tests, contracts) get enforced out-of-band, so the non-determinism stays bounded.
Ok so disclosure first I have made a product for exactly this problem.
But the answer isn't agents, skills although those do help it's an auto switching layer that chooses model automatically according to the goals.
But yes this tug of war is the reason I did build it.
Went down 40-50% tokens with much higher usage.
But this is headed to some kind of a reckoning. One comment about Gartner's hype cycle seems to be the best. In months the narrative has shifted from who needs devs to AI costs are uncontrollable. Both false but on opposite ends
But this is headed to some kind of a reckoning. One comment about Gartner's hype cycle seems to be the best. In months the narrative has shifted from who needs devs to AI costs are uncontrollable. Both false but on opposite ends
> Is anyone else going through this
I thought this became the norm already, isn't the whole industry working in the same way at the moment?
I thought this became the norm already, isn't the whole industry working in the same way at the moment?
You think not being allowed to write any code manually is the norm?
Only in the bleeding edge/kool-aid drinking bubble. I’m in a company that’s taking a gradual approach and supporting us with budget when we want to explore ways it might make us more efficient, but not setting any mandates. And we still have the expectation of understanding the output as if we’d written it.
My read is that we're traveling along Gartner's hype cycle. Tokenmaxxing/"who needs developers anymore" is the top of the peak. "oh no we ate our entire budget in 3 months" is heading towards the through of disillusionment. I wouldn't be surprised if at some point we hear about companies quietly walking back AI mandates and even gasp hiring developers again (not that it'll be shouted from the rooftops). And then, eventually, we'll figure out how to use these tools productively and settle in the plateau.
Different companies/industries move along the cycle at different speeds. E.g. the non-tech solopreneurs went from "I can vibecode an entire app in an afternoon, never hiring a dev again" to "whoopsie, Claude deleted my production database" pretty quickly (and loudly). Big tech also seems more advanced in the cycle; cf. Amazon's recent production problems and subsequent backlash, Uber's budget, etc. Larger companies are following the trend a beat late as their C-suite cosplays "we're a tech startup" on LinkedIn. And some companies are sitting this one out for the moment, either out of obliviousness or wisdom, and will adopt the tools once they've matured enough.
Anyway, give it some time; I'm sure the backlash will come to your company as well. Then, eventually, you'll be allowed to use AI rationally.
Different companies/industries move along the cycle at different speeds. E.g. the non-tech solopreneurs went from "I can vibecode an entire app in an afternoon, never hiring a dev again" to "whoopsie, Claude deleted my production database" pretty quickly (and loudly). Big tech also seems more advanced in the cycle; cf. Amazon's recent production problems and subsequent backlash, Uber's budget, etc. Larger companies are following the trend a beat late as their C-suite cosplays "we're a tech startup" on LinkedIn. And some companies are sitting this one out for the moment, either out of obliviousness or wisdom, and will adopt the tools once they've matured enough.
Anyway, give it some time; I'm sure the backlash will come to your company as well. Then, eventually, you'll be allowed to use AI rationally.
> "You shouldn't be manually writing any code".
Can someone explain how this mentality could possibly be rational? It sounds completely asinine to me, and I use AI quite a bit for my job.
In fact, the team at my work that seems to be 100% AI writes the worst code, follows no standards, and doesn't seem faster at all than teams that are simply AI-augmented.
Can someone explain how this mentality could possibly be rational? It sounds completely asinine to me, and I use AI quite a bit for my job.
In fact, the team at my work that seems to be 100% AI writes the worst code, follows no standards, and doesn't seem faster at all than teams that are simply AI-augmented.
Since then its been all agents, skills, MCP, harnesses, custom in-house frameworks, and running Opus 4.7 high non-stop.
Now at a company level, there are "encouraged attendance" workshops getting scheduled to learn how to optimize one's token use now that API pricing is becoming the norm at an enterprise level.
The direction I got from my direct leadership was very direct. AI / agents lead everything and the expectation is the team moves as quickly as it has. But the truth is most engineers candidly acknowledge: we don't fully understand anything. Especially because agents are also churning out the content of architectural documentation and user story requirements and acceptance criteria.
I feel like this is a situation where I am directly responsible for the non-deterministic output of these tools. The solution I get from my supervisor for any problem literally boils down to "You need to use an agent, skill, etc.".
Is anyone else going through this tug of war? How is it going?