Ask HN: What's your current AI agentic setup?

Pi running with Claude & Grok

Claude Code Ultra aka 20x Max, about $250/mo incl VAT that I have to pay

1) 10$ opencode go for personal work and a GitHub copilot at work 2) tried some llama stuff but otherwise nope. Don’t have the bandwidth to make it usable. 3) pi + herdr 4) no 5) no

Can’t say I need much more than this.

1. Two Claude Ultra, ChatGPT plus, Gemini through credits

2. Pay-per-token for work through Vercel AI Gateway (mainly same models + grok, but trying out others too)

3. Claude Code, Codex & OpenCode

4. Homebuilt https://svall.dev/

5. No

Wow I am pretty sure I read thru all comments and mignt be the first to say I use qwen 3.8 advanced because it's free

Google Antigravity CLI using Gemini Flash 3.8 most of the time. Amazing what it can do.

https://github.com/mkornreich/llm-desktop-electron with OpenRouter. I love Claude but Anthropic is expensive and too restrictive

For personal stuff, a $10/month OpenCode Go subscription with OpenCode, mostly running DeepSeek Flash. I've yet to run out of tokens.

Work provides an enterprise Claude subscription for work stuff, I've never hit any limits on that either.

1) I bought a z.ai 1 year subscription for their mid tier (second iteration of their pricing) for like 500€ total, so 40€/month for mostly GLM-5.3. I suppose flash is there too, but.. eh.

2) I have a 4090 which runs Qwen3.8-27b q4 fine with 180k ctx at q8. The model is capable enough to do things, but dumb enough to not be able to build elaborate nonsense like the bigger models love to do

3) I'm using https://github.com/juggler-ai/juggler which is alright. It could be a bit less vibecoded (and the readme and docs and all would benefit from less LLM writing), but, generally, the opinions shaping it seem decent, and I like just how much control over and insight into what happens it gives you.

I initially tried OpenCode, and, holy hell, what a mess. Do not ever use OpenCode. I also tried goose but it was too agentic for my taste.

4) No. I am doubtful that that kind of workflow can produce software I would deem worthy of production. I also personally just dread the whole overhead that is tickets, jira, etc, so recreating the same but now you're the manager seems like a personal hell to me.

5) No. I wouldn't know what for

People using DeepSeek - do you have a zero data retention provider based outside China?

1) Claude Ultra, Codex Plus, SuperGrok, Gemini Pro.

2) No.

3) Opus 5.5 on high for all the work. Reviewed by Codex by default. Additionally by Grok and Gemini (agy) for critical work.

I switched from Claude CLI to a custom-built harness because I got overwhelmed by walls of text in parallel sessions. Now I have a canvas per project where I spatially lay out cards. Each card spawns an agent in a worktree. The chat with it is condensed to WhatsApp style chat bubbles with all the thinking and tool calls folded away. The result of a card is a demo, usually a narrated video that I can comment on to trigger another iteration. Everything voice first, I mostly speak. Phone companion so I can keep working on the go.

I used to work with up to 6 Claude CLIs in parallel and couldn't go further. With my custom UX, I can do 20+ with fun. Now the bottleneck is actually my machine, so I added remote worker support to my harness to distribute the load to more machines.

(I open sourced the harness in the hope of sharing the fun. If you want to try it: https://github.com/sadilek/obeya)

4) No Paperclip or Munder Difflin or anything. I actually think it doesn't make sense to create persistent agent personas. Just spawn an agent when needed, give each full access to all skills and docs, and they'll figure it out.

5) Hermes Agent running, sending me regular news research reports, but frankly, it's just a cronjob and the whole Hermes setup isn't really adding value for me.

pi, a few custom extensions for pi, rex, gpt + ollama-cloud for a little ds4.1f

Muse and cursor. It beats every custom setup I had by far on ease of use. The cost is trivial.

Claude Code Ultra as the primary agent, Codex Pro as a second-opinion reviewer, with Rex as my terminal and multiplexer.

1) What AI subscriptions are you personally paying for? (Claude, ChatGPT/Codex, Cursor, etc)

No. I moved off of subscriptions earlier this year. I don't trust the subscription model from any frontier lab. Just look at what Anthropic has done. The first time I tripped a safety classifier on a seemingly irrelevant task, I canceled everything.

2) Are you using models outside (or instead) of those subscriptions? For example, pay-as-you-go APIs, OpenRouter, or locally hosted models.

Local first. I have a box running Muse Spark on my desk. My main inference box is a launch day DGX Spark (I saw the hardware shortage coming) and has my entire local AI stack. OpenWebUI and everything that feeds it. I moved away from SearXNG and started using SlopSearX (it's quite good).

Everything else goes to Openrouter. I use few models but GPT-6 Luna is my go-to, sometimes I switch it up. Really depends on the task I need to perform.

3) What coding harnesses and interfaces are you using? Are you sticking with the default CLIs or IDE integrations, or using alternatives like OpenCode, Pi, Aider, Muse Code, or custom-built harnesses?

Github Copilot for work. It's meh. Personal projects and general personal use? I started with OpenCode and have since migrated to Pi (omp specifically).

4) Are you using any agent orchestration or swarm tools?

I thought about it. I haven't really felt like I would derive value from changing my workflow and tools. It took me a while to find a good mix of utility, control, and efficiency and haven't wanted to change it until at least next year.

5) Are you using broader agent platforms like OpenClaw / Hermes Agent / Grok bot / Dots / Muse?

I have a dedicated Mac Mini to run Hermes. It is pretty much integrated into my entire life now. I could live without it, but automating general purpose things is the entire point of why I put that machine together.

How is this different from this time last year? Radically different to say the least. The pace of change has been EXTRAORDINARY to say the least.

I'm using Claude Code (Ultra) across a multitude of tmux sessions, over two or three laptops – largely so long running actions don't happen on the actual laptop I take with me (nothing worse than waiting around for an agent to finish before you can leave a café). To make it a little easier to visualise across those sessions I built https://belfry.robgough.net which is like herdr in some ways, but just uses standard tmux.

I also gave in after seeing my co-founder use it, and gave GrokBot a go. I use this for everything that's not coding or design... so marketing, email reading, right now it's currently inputting my meter readings to my electricity company after emailing the building management company and getting photos back. It's super useful at little chores that eat up time. I have it daily check the status of my various projects and services. I love it tbh. I've also built my own version, which I might move to... though now it's all set up it's quite sticky, and it already spotted an error my accountant made on some taxes so it's already saving me money and I'm feeling quite positive about it.

I'm getting a bit frustrated with having to run things on my laptops though -- it feels a little pets not cattle, and wish it was all in the cloud. I'm toying with building out an agent factory, but frankly there's no super-cheap way of doing this so I'm holding off, though I can see me building a tool that connects those laptops to it creating sandboxes with Apple's `container` (which I'm using successfully on my grokbot replacement).

Mainly CC Max with some API calls to OpenAI for different things.

No harnesses right now. Ok occasion using models in LocalLM.

Haven't got into agent orchestration much other than Paperclip earlier this year.

Claude desktop and mobile, but only because I am wary of installing any software from new vendors right due to widespread security vulnerabilities in all software.

1: I have the fancy Claude, and then regular paid Codex, Grok, and GLM 5.3

2: Not for development

3: I work inside Claude Code almost all of the time, inside a byre sandbox, often connected to happier. Claude calls out for code reviews to the other models

4: I have a self-built code-review wrapper for Claude to call the other models, but nothing more than that

5: Haven't managed to find anything useful to do with them.

1) 20$ Claude subscription. I used to have 2 20$ subscriptions but not anymore.

2) My main model is Deepseek flash via the API.

3) Claude Code CLI inside OpenShell. I like Claude Code's UX. Lots of custom skills and scripts.

4) No

5) No

Claude max; Atomic; Jira; herdr and VPS

very simple; claude code on pro subscription acting as a project manager, deepseek acting as developer

1) None. duck.ai for random questions/research for private use. No personal coding project or anything like that.

2) Yes, Anthropic models usually, at work. Pay as you go API. Company is trialing limits, routers and open weight and even local models. Since the Anthropic 5.5 models none seem reasonable to use any longer (for now at least)

3) Claude Code at work. Tried Aider early on and it failed to even read its initial context (Claude models like a year+ ago). I dropped it dead after that - unusable if you need me to specifically "load" and point our files. That's not agentic coding, sorry. Have not used the actual features my IntelliJ Ultimate subscription provides in quite a long time, though I loathe to tell IT/finance that. I do use it from time to time to double check what Claude tells me. Used `pi` w/ local models (some gemma and qwen that'd run on my hardware. nothing usable w/ the work code base - or much anything for that matter, like I'd not even get through an interview coding challenge with that if you ask me)

Did use Codex maybe like 8 months or so ago to ping pong implementation plans and reviews between GPT and Anthropic models but then just stuck with Anthropic for some reason.

4) Nope.

5) Nope. But we do have our own in-house review agent and software factory agent. Like think CodeRabbit et. al.

1. Claude Ultra, ChatGPT Pro 200, SuperGrok, Kimi Pro, Cursor Pro, Google AI Pro, Muse Code High Usage. It comes out to ~$500/mo, mostly from Claude+ChatGPT

2. I use OpenAI/Anthropic APIs for various workloads (app features, automations, etc.), but not for personal inference/coding. Recently both of these providers started giving $200/mo in API credits, and OpenAI already gives free tokens, so I don't usually pay for this usage.

3. I always use the provider's preferred CLI: Claude Code, Codex CLI, Grok CLI, Kimi CLI, Antigravity CLI, Muse CLI. I use this with herdr to manage everything, and then a small custom app to keep track of my usages since I hit limits quite often.

4. No. Just the CLIs. For very large projects I've found Claude Code to be the best -- it can very easily manage many subagents.

5. I've tried OpenClaw, Hermes, Dots, and Muse. I really just haven't found great use cases for these tools

I use Claude & Codex for most of my personal coding/projects. For the other providers I mostly want to have broad experience with different models/harnesses. It's very interesting to see how each performs and what kind of tasks they do/don't excel out. The only con of having cheap plans is you really can only give them simple tasks.

---

At work I use AWS Bedrock w/ Anthropic + OpenAI models at API pricing. Each employee is given a rolling 28-day budget w/o access to Fable/Astra. I use the same Claude Code/Codex CLI + Herdr.

1/2) OpenAI Business Premium subscription and a locally hosted Qwen 3.8 Flash Next running on a DGX Spark.

3) I use both Opencode and Pi.

My use cases are mostly analytical work, document generation, and code generation and review.

I used to have the Pro subcription for Claude but switched recently to the equivalent Plus subscription for ChatGPT. With Anthropic, you cannot access their inference servers without going through one their client software or paying the API pricing. It is really annoying as those are two different product categories and I'd like to use the client of my choice with an LLM. OpenAI allows you to do that, so I have a proxy running on my computer which connects to OpenAI's servers in the back ground and serves locally a Chat Completions-compliant API, with the requests counting towards my subscription.

I also have an OpenRouter account, I use it regularly but mostly to explore new models and see what's out there. I wish I used it more but the bigger part of my requests, almost all of them, go through my subscription. My thinking is since I'm already paying for it, might as well use it. Which is a shame because it prevents me from exploring more.

I even think a combination of the free tier of Big AI with a pay-as-you-go pricing for more complex requests would actually be cheaper for most people than their current subscription, even with their heavily subsidized pricing. The problem is that most people wouldn't want to think of when they should go deeper with a bigger model and wouldn't want the mental task of switching. The same way people don't really switch thinking levels with a given model, even though it would be a more optimal use of their token pool.

There is a huge market for an alternative business model around LLM inference. The big providers will obviously go towards an ad-based model in the following years, which will be good enough for the greater public, but a lot of people and small companies would be interested in packages and subscriptions that suit their specific needs and lines of work. OpenRouter has a great product and there is so much possiblity in this space. I also believe there is so much potential in the development of models based on data primitives other than text or rasterized images, which would be really useful in a lot of industries.

For the frontend, I either use the web-based version of ChatGPT for quick questions and analytical work (the desktop version is unnecessarily heavy), or Open WebUI. For coding, I mostly use Zed. I like the traditional development environment and tooling, I tried several agent-first environments, but in the end I prefer an IDE on top of which agents were added. OpenCode Web has evolved a lot, I have been using it for quick scripting as well. I never use any coding agent TUI anymore. Behind the scenes, the agent harnesses are OpenCode and Codex. I use them interchangeably, I don't see much of a difference. Zed connects to OpenCode and Codex through ACP. Both OpenCode and Open WebUI connects to OpenRouter and the OpenAI Proxy. Open WebUI has Open Terminal, which I have been meaning to explore, for there to be less friction between any analytical work and a document that would be needed to be produced out of it. I also run small models through llama.app from time to time, when I want to be reminded how weak my computer is.

I don't use any orchestration tool. I don't have any work that involves so much volume that it would require that kind of setup. It's more important to me to maintain the right mental model, I don't want to risk any discrepancies between what I'm thinking and what's done. Therefore the agentic work must be single-threaded, if that makes sense. I also don't use AI chat gateways like OpenClaw or Muse because I don't want to do that work on my phone every minute of my life. I'm either working or I'm not.

This comment is longer than expected. It's such a fascinating and still evolving space.

VSCode + OpenRouter + flavor of the day cheap model.

I spend about $5 a month with this setup.

We built our own harness local & cloud models, test it out: www.vekmem.com

Literally copy pasting prompts between ChatGPT and claude code all day every day.

I skim, push back / ask for changes to prompts if/when needed (and very, very rarely edit any prompt myself, and never edit any code myself).

I jockey about 1-5 of these ChatGPT<-->cc sessions, my sweet spot is 2.

This is like the stone age, but I get as much (or as little) awareness as I need and generally the level of control that I need for whatever happening. And it's easy as heck, very little load, just how fast I can read or type. Sometimes I voice dictate, but I type decently fast.

For things needing less of me as an expert (my real value, and where I try to say), I can just say "do thing <7-10-word description>", obviously.

None.

1) I mostly use Claude Code Max (paid for by my employer) and ChatGPT Pro (which I pay personally)

2) I use Claude for all work and programming, ChatGPT for random questions in my day-to-day life.

3) I just use it through the CLI as it's my favorite interface, often with multiple CLIs running at the same time

4) Nope

5) Nope

so much inspiration in here!

hard to contain myself not to change my setup too often (instead of doing actual work)

Im curious how hardware/firmware engineers are using Claude, et. al. for work. Tools/MCPs that are worth it?

Only GitHub Copilorpt for documentation. Sponsored so cost nothing.

1. Claude Ultra + a Codex Plus. Probably going to drop Codex soon, GLM is gradually taking on what I intended it for.

2. GLM-5.3 and OpenCode for cybersecurity

3. Primarily custom built harness. Rust centric, heavy on the production engineering gates and a bunch of graph networks for context control.

4. Custom.

5. No.

My general ethos is that if taste is what is important going forward, my code generation must have it's own taste. If you can download my tooling off the shelf, I am replaceable by anyone who can do the same. My end goal is to drop Claude Code once the harness prices start to drift to API pricing. In the meantime, anything that can be automated into traditional and efficient code will be one more thing in my control regardless of future token prices. When I hit my limits for the week, I spend the remainder of the waiting time going over the framework data to see what are my largest sources of consumption and iterate in automating that more.

None. These things aren’t that good.

1. No

2. OpenRouter PAYG - glm or opus for most things

3. OMP or CC

4. Zellij/Vim — some things never change

5. No

Claude App + Opus 5.5

Codex on Pro 100, Claude on Pro 20.

I don't even understand how people can spend thousands of dollars a month on LLM tokens since I've accomplished multiple major projects with this setup. In general, Claude's better at frontend and debug and Codex is better at feature implementation and research from what I've found, which is why I have both.

2. No. GPT 6.1 Sol is very token efficient that I pretty much never run out before the end of the week anyways.

3. Default Codex/Claude app. I only use VS/VSCode to hand-edit/git sync things when needed.

4. No. Swarm is a waste of token imo, and a single good human in the loop beats any automated orchestration tool.

5. I find it unnecessary.

Claude Code. Everything else default!!!

1) OpenCode Go $10, though I'm just experimenting with it. Before I have just been using the free models available in OpenCode, since sometimes its hard to justify a subscription when you're unemployed.

2) I tried a local Qwen model but my poor old 1660 Ti did not enjoy it, and it wasn't all that useful.

3) I primarily use OpenCode running in a podman container for actually working on stuff. Also use the default free chats with Gemini 3.5 Flash for general searching, ChatGPT for brainstorming, and Claude to answer more code related things if they come up.

4) Nope.

5) Not at the moment. I've been eyeing using Hermes, but since I don't feel comfortable paying much at the moment I haven't really tried.

My own harness (nothing intricate, but to my taste), DSv4 1017 on my own hardware

Keeping it simple.

1. Claude Max

2. No (only the API for projects)

3. Buzzpot.app (my app)

4. No

5. No (I tried them, but couldn't find a use case for them)

I mostly made my own claude messaging system and then context engineer the fuck out of it by thinking from first principles. I never needed skills for example because I already had something working before they were invented

Codex (Pro x 1), no outside models, custom harness, custom orchestration, custom agent harness.

I avoided building anything custom until I had to, and I avoided adding more than a single agent until I had to.

My preferred way to use AI for coding has been to work closely with a single coding agent with the code open. That changed wth 5.6-Sol and a handful of skill files. I stopped needing to watch closely to keep it from fucking up the codebase. With the right request, that combination will get the job done 9/10 times.

That unlocked a new workflow: as soon as a change is requested, a coding agent reads it and gives me a PR. I want a complete proposed code change in front of my eyeballs by the time I open the change request myself.

So I have a custom solution in place to manage a handful of agents aligned with this workflow.

Custom solutions are so cheap and easy now... it's really hard to put up with solutions that are only mostly right for the job.

1. Using ChatGPT with a Business Premium seat (not personally paying for it).

2. Also have some API keys for some open weight models in Europe, but currently not using those.

3. I've personally used OpenCode, Claude Code and Codex and of those, I liked Codex the most, till bugs kept on accumulating and they started doing all kinds of shit like that daemon recently. I now switched to https://github.com/capocasa/3code which is a small, simple agent harness but with built-in sandbox, unlike so many other simple harnesses. I really like this one, plus the author is very responsive to feedback

4. No. And I don't think that stuff works in the long run with current quality of code output

5. No.

Dont use it at all. Seems like a waste of time for what I do as far as I can tell.

2) I am running Qwen Flash Next on Strix Halo with Halogen inference server. This is great setup that can get you 4 concurrent agents for 300W of power with 1k prefill and 30-80tk/s decode. I do not fill like using subs that I have from work with this.

1) Xiaomi Mimo Pro 2) I purchase periodic credits through Tailscale Aperture 3) MiMocode for working with Mimo, Pi with the others 4/5) Nope