Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

AI coding at home without going broke

stephen.bochinski.dev

Post reply on HN
Print view View original

201–210 of 321 posts

Re: AI coding at home without going broke

#201

Post by gaflo »

tunesmith wrote: post #47

I feel like I must have plateued and don't know what to do next to level up. I'm currently on the $100/month codex plan and it seems fine using 5.5-xhigh all the time. I think of what to do next, have a chat session to determine exactly what to ask for up to the point of being ready to implement, and then codex churns on a commit-sized task whereupon I briefly check it on my local dev server. If necessary I ask for a…

Can I ask what exactly you are building? Your experience tracks for me when building a real product -- something I want other people to use. Most of my time on these projects is spent talking to my users and carefully refining my requirements and design.

For personal pet projects I can definitely see how you can blow through your token budget very quickly. If I just point my coding agent to iteratively come up with some heuristics for some NP-hard problem, it will read intermediary outputs and constantly make small changes "in the dark" until it either finds a small improvement or gives up. In a similar vein I found that you can burn many many tokens if you try to let the agent reverse engineer something where you don't have the source code. If you just give it a binary or some interface to work with and a vague task you can easily burn your entire budget with 1 prompt.

I wouldn't want anyone to use these fully vibe coded toy projects though; it is more of an exploratory curiosity for me where I learn more about some problems I'm interested in as well as gauge how good the agents are at tasks that I seem to have a much better intuition on how to approach.

Re: AI coding at home without going broke

#202

Post by bthornbury »

tunesmith wrote: post #47

I feel like I must have plateued and don't know what to do next to level up. I'm currently on the $100/month codex plan and it seems fine using 5.5-xhigh all the time. I think of what to do next, have a chat session to determine exactly what to ask for up to the point of being ready to implement, and then codex churns on a commit-sized task whereupon I briefly check it on my local dev server. If necessary I ask for a…

promote yourself to PM only and use agents for authoring, verification, tests, checking the tests

orchestrator -> parallel subagents with investigation, authoring, verification, benchmarking subagents and integration / final verification handled by parent has improved my productivity too.

I feel like from here its agent swarms against a whole spec but haven't got there yet.

Still getting plenty of bugs in the more complex scenarios, but mostly (in some projects) i never have to look at the code and treat it like a black box

Re: AI coding at home without going broke

#203

Post by ryeguy »

alecco wrote: post #181
Yoric wrote: post #78

Earlier quoted context omitted.

Directly at DeepSeek? It was my understanding (but I didn't check) that some other AI operators were providing (some of?) DeepSeek's model for cheaper prices. Still, that's interesting. What do you get for that price? Only coding, or also e.g. image generation?

DeepSeek API gave 6x to 8x better caching rate for inputs over OpenRouter (even chosing DeepSeek as provider). And some of the cheaper providers are using FP4 quantizations. https://openrouter.ai/deepseek/deepseek-v4-flash-20260423#pr... After complaints the cached read is not listed anymore in that page, you have to click one by one. All providers for DeepSeek V4 Flash charge ~$0.02 while DeepSeek provider is $0.002…

Openrouter's pricing via the deepseek provider is the same as the official deepseek api for both flash and pro and for cached and uncached tokens. It's literally the same api.

And no, cache rates are not different if you're going through the official deepseek provider. The only way caching rates can drop is if you let openrouter fully control routing by preferring uptime or something, and then it might bounce you between providers. But you can control which providers for a given model are in its routing pool and stop that.

Re: AI coding at home without going broke

#204

Post by anon373839 »

mwcampbell wrote: post #33

I invested about $4,000 in an NVIDIA DGX Spark several months ago. 128 GB of unified RAM, and the NVIDIA GB10 chip. With the RAM, the several CPU cores, and the 4 TB NVMe SSD, it's a very capable ARM64 Linux computer even without the GPU, and so far I've mostly been using it as such. But I wonder, what's the most capable model, specifically for coding, that can run well on that hardware?

Qwen 3.5 122B can fit with context at a pretty high quant (Q6). That's an excellent model.

Re: AI coding at home without going broke

#205

Post by lee_ars »

mwcampbell wrote: post #33

I invested about $4,000 in an NVIDIA DGX Spark several months ago. 128 GB of unified RAM, and the NVIDIA GB10 chip. With the RAM, the several CPU cores, and the 4 TB NVMe SSD, it's a very capable ARM64 Linux computer even without the GPU, and so far I've mostly been using it as such. But I wonder, what's the most capable model, specifically for coding, that can run well on that hardware?

I'm currently working through research and testing for an article on Ars about the Spark and what things one might do with it, and I've kind of stumbled into a two-LLM agentic setup with Qwen3.6-35B-A3B (via nvidia/Qwen3.6-35B-A3B-NVFP4) as the planning agent and the FP8 version of Qwen3-Coder-30B-A3B-Instruct (Qwen/Qwen3-Coder-30B-A3B-Instruct-FP8) as the coding agent that the planner delegates tasks down to. I'm sticking with vLLM as the inference engine, and I've got it wired together into a 2-agent loop with Opencode.

The Qwen3.6-35B-A3B planner hums along at 50-55 tokens/s, and the Qwen3-Coder-30B-A3B-Instruct coder does 30-35. With both agents up and ready to work, RAM consumption sits at about 112 of 128GB.

It's pretty okay. I'm faffing around with having it disassemble old MS-DOS games from the 1980s, which is a task that lends itself well to the setup. It's not the fastest thing in the world, but with the planner's context window at 256k tokens and the coding agent at 128k, they chew through pretty long task lists handing things back and forth without complaint. The only real issue is that even with really tightly scoped prompts, the coding agent tends to hallucinate like it's on LSD. But the planning agent appears to be quite good at spotting the hallucinations and re-parceling work back to the coder.

It's neat. I'm going to be sad when I have to return the review unit in a couple of months.

edit - I also have been fiddling with Deepseek v4 Flash via Antirez's setup (https://github.com/antirez/ds4), and it's pretty fantastic (and fantastically easy to get running). It's pretty pokey on the Spark, though, at 14-ish tokens/sec. And unless you have a second Spark, it's going to be the only model you run at one time, as it eats alllll the rams.

Re: AI coding at home without going broke

#206

Post by ynxshiny »

12_throw_away wrote: post #199
dozerly wrote: post #176

Earlier quoted context omitted.

We are so many layers deep in AI hype that I honestly can’t tell if this is /s or not

"Make no mistakes" is I thought a phrase used to make fun of "prompt engineering," not something people really do?

"Claude make me 1 million by tomorrow, no mistakes"

Re: AI coding at home without going broke

#207

Post by efavdb »

12_throw_away wrote: post #199
dozerly wrote: post #176

Earlier quoted context omitted.

We are so many layers deep in AI hype that I honestly can’t tell if this is /s or not

"Make no mistakes" is I thought a phrase used to make fun of "prompt engineering," not something people really do?

Pleading has worked for me. “My job depends on this, please help me” and ChatGPT would do a task it previously claimed it wasn’t able to (extract text from an image, it claimed it couldn’t make it out at first)

Re: AI coding at home without going broke

#208

Post by dofm »

apsurd wrote: post #182
dofm wrote: post #165

Earlier quoted context omitted.

I must say I am not quite just venting. I have been struggling severely with burnout for a couple of years and as I work to fix it by myself ultimately, and get back who I was, the awful thing is finding out that the industry is so utterly and completely different anyway. So in my fight back I decided that I needed to re-centre myself; learn how these tools can help me personally return to productivity, try to get th…

I'm younger, but not by much and I too feel instinctively sad by how abruptly the entire industry has changed. And there's no going back. It's because I'm a craftsmen, I care about the code. And you learn in your career that it's a bad idea to care about the code, especially in a business context, which one's career is very much trapped in the business context. I care about the code because the code is the product in…

> I think you're product person caught in backend-dev circles.

I am kind of all the things (product design, dev, front end, training) because at the small end of things you have to be; you don't get directly paid for misery-avoidance but I don't think that's any reason not to do it :-)

But thank you.

Re: AI coding at home without going broke

#209

Post by dottchen »

running 2 $200 codex subs seems to work for me. It's quite easy to run out of a full account's weekly usage if using xhigh and fast mode all the way, and i'm not using it for autonomous running, still mainly human reviewed actual work.

Re: AI coding at home without going broke

#210

Post by kapperchino »

tunesmith wrote: post #47

I feel like I must have plateued and don't know what to do next to level up. I'm currently on the $100/month codex plan and it seems fine using 5.5-xhigh all the time. I think of what to do next, have a chat session to determine exactly what to ask for up to the point of being ready to implement, and then codex churns on a commit-sized task whereupon I briefly check it on my local dev server. If necessary I ask for a…

On the topic of access control, I’m building a coding agent with no shell access, currently only supports rust though. https://github.com/Kapperchino/agent-joe
Post reply on HN