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

Claude Pro. Mainly used for developing my apps Pagecord and Feedgrab, but also for personal stuff like diy problem solving. travel planning, data chores, etc.

I feel I get a wild amount of value out of £90/mo (paid for via my apps revenue)

A key part of mine is using voice input to give max context quickly to my agents. I know some solutions exist, but on Linux they all frankly suck

So I made my own. Including patching onnx myself to run parakeet model (way better than whisper)

Fully FOSS: voiceio.dev

1. Paying for Open Router only.

2. Local VPS: serve small models like Desert Ant, Laya, etc only (the VPS has GPU).

3. Hermes only: Stick with one, arch is converging.

4. No. Used to use Hedr but that's about it. I ask Hermes to do that for me. (or create custom Hermes agents by specifying graph in text)

5. Just Hermes: refer to #3. - Fun at home, build, automate, disabled gateways for online chatting becauese of bugs (not sharing)

Hermes, OpenRouter, . No Claude/Codex sub too expensive. Use them at work instead.

I pay for Claude Pro ($20/mo)—using Claude Code with Opus 5.5, and Google AI Premium ($20/mo) for Gemini Pro with Gemini 4.

When I write code with one, I use the other for peer review.

I feel like this is a great opportunity to mention https://github.com/regnull/openbot, my attempt to create an open-source tool similar to Grok Bot. You can hook it up to OpenRouter and solve 80% of problems with 1/10th of the cost. I use it all the time. Contributors are always welcome.

I'm using claude with a pro subscription. Claude Code extension in VSCode devcontainers and Claude CLI in a restricted Windows account in powershell terminal for android development. No orchestration other than a directive in CLAUDE.md to allocate coding tasks to sub-agents.

I've moved all my ssh keys into a key-agent which prompts for key use as I got alarmed by claude attempting to login to remote systems to verify things as it worked through problems.

I was using VSCode with Claude Code extension, multiple tabs at once, multiple instances per worktree.... then I listened to the Lex Fridman episode with DHH [1] and they mentioned Herdr [2] and I tried it out - NEVER LOOKED BACK! It has been an extremely long time since I completely abandoned and changed my setup - surprised myself.

Herdr is awesome - it pings me when an agent is done or needs my attention and shows me all my terminal panes that are running an agent that I can access with one click.

I am now 90% Herdr terminal, open IDE maybe 1-2 times a day just for more convenient way to look at something. I searched ways to stay in the terminal - found lazygit and love it!

I use long-running worktrees (e.g. named worktree-1, worktree-2, worktree-3, ...) and have parallel work going on in them so I can work on whatever I want.

In each Herdr workspace, I have three tabs: 1st is for ~3 panes, one for Claude and two or more to run the app and basic terminal. Tab 2 for lazy git, and Tab 3 for an extra terminal when I need one. For example, right now I have ~20 Herdr workspaces open now between a few projects, multiple worktrees per project and a few across remote machines ready in case I need them. When in the zone, I might have 1-8+ agents working on completely different tasks.

I have a couple remote machines over SSH that I use when I need a task to run and I need to close my laptop.

I have a keyboard shortcut to init my workspace to that setup. lazygit is amazing.

I ask claude to do everything for me, for example: "Let's do it, open a jira ticket, assign to me and put into sprint. Fix it and open a pr"

I create artifacts all the time to share findings with team or as a checklist of manual testing I need to do.

I find that gh copilit does pretty good review on PRs, and I have a claude skill to go address all the copilot feedback (review, validate, reply, resolve, push changes if any were made).

I was on the 200/mo Claude plan but had to switch to the Team plan, so I pay 120/mo for two users and I swap to the 20 plan when I hit my 5hr limit on my 100/mo plan.

[1]: https://lexfridman.com/podcast -> DHH: Future of Programming, AI, Agentic Engineering, Vibe Coding & Linux [2]: https://herdr.dev/

bit late of a reply, but I run this simple stack for some time now, and I'm enjoying it quite well: https://stack.yiin.lt

Until local LLMs become feasible on my modest hardware, I will be keeping AI tools safely confined to the browser. Never will I grant any kind of admin permission to black-box software that phones home to an oligarchic corporation.

In practice that means Duck.AI and Perplexity.

And the system clipboard.

i use hermes-agent as my personal assistant and it talks to my claude code and codex if it needs to do projects and other stuff. openrouter is my default for the hermes-agent and sometimes i connect to lm-studio if openrouter gets errors or maybe try out some local models

1) None

2) No

3) N/A

4) No

5) No

1. Claude Ultra

2. No

3. Claude Code

4. My own engineering harness: https://devarch.ai/

5. No

I started using ChatGPT in the beginning then switched to Claude then Claude Code then developed DevArch over that entire time.

1) I pay for claude ultra coz the plan rates are significantly better than the api rates, and i feel like i can tokenmax harder than if i am constantly watching my credit balances.

2) yea openrouter free models, usually nemotron coz they have decent availability and I don't get constantly rate limited like with the gemma models on there.

3) honestly still using the default IDE, used to use claude code cli but nowadays i just launch claude remote control and just talk to it through my phone.

4) something custom, i wired it all up in slack so i could easily talk to agents through the same place i do my work

5) Used to use openclaw on a raspberry pi to experiment with like autonomous development but i needed something with a bit more transparency so i moved everything into slack.

Claude Code Ultra in Emacs. Split Emacs windows: Bottom is Claude Code, top left is code, top right is db-backed editable spec (propositions and premises) with Clojure like eval that sends spec to Claude Code.

1. Claude Ultra for personal, Claude Max for work, Codex $100/mo

2. OpenRouter, DeepSeek, Luna, and testing out models here and there

3. Claude Code and Codex for the respective subscriptions. Played with OpenCode and Pi some as well

4. I've tried a number of these tools and settled on my own tool (not released) to oversee my agents

5. Played with OpenClaw and Hermes in the past, cool ideas but execution fell flat. I currently use Dots and Duet for little things (mostly monitoring communications). Dots is much better than Duet today but I hope Duet gets better.

1. none 2. nvidia nim, opencode zen, google free models. local QFN iq3_xss on m1 max 64gb, paperniuk splash 3. pi agent, pi pocket 4. no 5. no

1）claude（pro）、ChatGPT（business）、Kimi（pro）；各一个订阅； 2）是的；Deepseek（API）； 3）默认的CLI； 4）否； 5）否。

我是谁：一名做基础医学数据分析的博士研究生。

1) $100 Anthropic via work 2) codex personally 3) Claude Code + Herdr with a root agent steering everything. Waiting for Rex 4 and 5) absolutely not lol

im midday through building a 3rd party multi device coding and everything harness with some unique hard to replicate features that should really help for successfully building amazing stuff reliably.

despite some of the stuff ill need to do as a consequence of wanting to keep some of those tricks private necessarily requiring that i kinda have my own byok open router for zdr providers endpoints only, i think its worth it.

theres a lot more i could say, but i hoping it can wind up being a cult favorite in techy/professional circles even if im planning to do something like 50/month for core plus pass through with hair cut on usage based costs. cause despite the large theoretical market, i think taking a page from sublime text for product stuff is likely the best way to roll.

i just want nice tools and recurrent income thats stable and mine. if both groove together, thats the dream.

1. OpenAI and Anthropic, both at their 100 USD tiers, one account each. Would do 200 USD but I need the money to help a friend instead.

2. Previously yes (esp. to try out new models on OpenRouter, like GLM 5.3 or Kimi K3.1 Pro and the various DeepSeek ones, Mistral ones etc.) before also having their subscriptions for a bit. Wrote about them on my blog, seemed like nice alternatives to OpenAI and Ant, but didn't give me as many tokens (and that's with the BS those two are occasionally pulling).

3. Codex and Claude Code, for a while had a custom launcher for Claude Code (ccode) for 3rd party providers, also tried out OpenCode (really nice CLI, but barebones web UI/desktop versions), alongside additional ADEs like Paseo (really pleasant, but sub-agent status updates hanged sometimes) and Kepler (cool ticketing system integration, but the UI wasn't very pleasant), before just going back to the two in the end. Same as with MCPs and other stuff, realized I just don't need most of them, near-SOTA models are good enough to automate long form work and work in parallel well enough, all you need are Git worktrees and a decent OS environment to work in. Still waiting for some dumb agent to wipe my drive though.

4. No, closest was Paseo or Kepler, but those tools would help me more rather than beat the agents into shape. The most I do orchestration wise is demand adversarial reviews (main session often spawns sub-agents for implementation and then has sub-agents with clear context for review in a loop, if issues are found, they're solved, then back to review, until nothing too serious remains), seems like even SOTA models fuck up consistently enough that they benefit from such reviews still. Meanwhile the most I do in regards to work is nudge how to best do exploration and division of work, when I know that something will need a bunch of parallelism or not to pollute the main context with bunches of side quests.

5. No. Maybe I should, but given how often LLMs lack decent judgement when the tasks are underspecified, I don't think I'd benefit a lot. My ALL-CAPS-SWEARING/SESSION metric is still higher than I'd expect from a reliable assistant system, despite agent instructions and memories and custom tools and linters and all that. In a word, even SOTA models still do stupid shit and bog down workable designs with complexities and mechanisms and concerns that nobody fucking needs. Computer use is pretty good though.

At work, we're using Kiro. I've got no complaints.

In my personal life, I'm extensively using Hermes. If you look at some previous comments you can see what I use it for. I genuinely think personal AI agents are the future, the way I feel when I use it is how I felt when I tried out the first iPhone. In 5 years time, I think most people will have a personal AI agent in the same way the most people now have an e-mail address.

1. Claude Max - two (personal and work), ChatGPT plus, OpenAI (PAYG at work), Gemini PAYG (mostly for automated transcription). I could probably cut my personal Claude Max account, but hang onto it for the constant stream of personal project ideations.

2. Yes, per above

3. Mostly Claude Code, occasionally OpenCode and Codex. Rex/Ghostty as the interface to it all.

4. Yes, lightly, a custom orchestrator

5. Not actively, have tried most ex Grok Bot, but haven't taken the time

1) Claude Code (Claude Teams Premium seat) plus pi.dev (self-hosted GLM-5.3) at work, Claude Pro at home. Opus 5.5 Extra effort for ~99% of the prompts.

2) Not really, messed around with Openrouter and Deepseek for some stuff that Claude refused to do

3) I use Claude Desktop (Linux) and a heavily modified pi.dev at work VSCode as IDE. At home just use the iOS app and Claude Desktop (Windows).

4) No

5) I have Hermes Agent setup at work and home but don't really use it.

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.

At work Claude Desktop with Opus 5.5. At home Deepseek Harness, 5-15USD per month. I don't like CLI harnesses. coz it looks too monotonous. I prefer default agent orchestration of the harness, since its their job to take care of it. not something I should spend my time and efforts on. If I did, high chance it will get out-dated soon. I do not trust agent platforms. Lethal Trifecta comes to mind.

I am currently stacking 5 opencode go accounts, spending $50 a month for $500 in deepseek v4.1 flash credits, using OMP as my harness

1. 20$ Anthropic

2. deepseek api, local qwen/gemma on fw desktop

3. Claude & Pi

4. No 5. No

I'm basic "type what you want in prompt, read what comes out"

1) the cheapest Claude sub, $20usd I think.

2) pretty much just opus

3) claude code - but with Orca + app (being able to dev remotely via the app has been great for personal projects)

Work has me stuck with Cursor, which I don't find as good, even when using claude models.

1. I don't pay for any subscriptions personally. 2. I use locally hosted models, Qwen 3.8 Coder 3. I use VSCode, I used Opencode for a while but VSCode then supported OpenAI compatible so I use that with ollama.cpp and Qwen locally. 4. Custom build orchestrator that I built as part of my learning with AI. 5. Nope.

Mostly https://tix.im/ with my vx-maintainer expert, codex, opencode and claude-code as harnesses.

I was unhappy with everything out there - not secure enough, not flexible enough, not composable enough, not nix enough. The whole concept how we use AI is garbage, having 10 terminals open with agents working in different branches. I build workflows, very complex, multi model, multi harness, parameterized - nix defined workflows that run in highly isolated, fully composed environments. I have a few dozen workflows already, in vx-maintainer: quick-fix, change-management, spec-implement, red-green, sync-branches, ....

Each of those workflows does something, and this extremely well. Every time a workflow fails, the workflow gets some massage to make it more resilient, more flexible, more correct. Workflows are usually design that restart just recovers the last state and continues work. My land-feature workflow has at least 40 steps, most of them deterministic scripts. For example, if rebase is true, a mechanical rebase is started, if this fails, an agent rebase agent is started. Full multi panel + review tool pipeline. With time, the workflows start to be super sophisticated, for example: review is cached, but the review is pinned to the commit hash. When restarted and rebase is on, a ast-diff is generated to check if the review can be reused. Stuff, no normal person would do :)

vx-maintainer has an own bug / change tracker. Normal bugtrackers are build for humans, such agentic system need a tracker that works for both.

Then I have a special agent, the queue-manager, his goal is to keep the workflow engine busy. He queries the vx-meta tracker for changes and workflows to get implemented.

Workflows are semi-automatic. You can have tty and ui programs running inside the workflows. You have decision steps and agents can ask the operator.

And this is just the surface of it.

I never was able to get so much stuff done and had so much fun getting a project build. And fucking safe - no more fear when updating dependencies.

I use Orca as an “ADE” (mostly for the multi-worktree interface and mobile app).

In my day job I use Claude Code & Codex subscriptions (Codex is mostly for review, second opinion, and as a backup).

Personally I use both oh-my-pi and Hermes with a Codex subscription. I tend to use OMP more now though, since I find its interface a bit nicer when working on the laptop.

two max Claude suscriptions, active dev in vscode to share screenshots, GitHub tickets for memory, basic skills, superpowers,subagentic coding with opus and the main in fable, daily Cron tasks for answering email/linkedin/etc. Desktop Claude for video/music editing . No API usage, expensive for my needs.

I’ve moved from paid subscriptions on Codex to Claude to Muse.

I now use Muse as my primary agent and then spin up and manage Codex and Claude agents as needed on an Oracle server.

Pi running with Claude & Grok

For now, I have decided to refrain from using LLMs because I consider the technology a very serious threat to humanity. So my answers are:

1) None

2) No

3) None

4) No

5) No

After trying so many different setups, now I just use Claude-- the only thing interesting about my setup though is that I separated a "work-claude" and a "personal-claude" install on my laptop, which isn't officially allowed and took a bit of wrangling. But I basically had to do it legally.

1) no personally paid - employer gives me claude 2) no 3) corgi + claude at work 4) no 5) no

1. Claude pro monthly

2. Kimi on OpenRouter, Abliterated.ai as needed

3. Claude Code (with all models), Claude Desktop, tried pi.dev, opencode

4. No

5. No

Main use case(s):

- Software applications

- One-off automations (calendar sync from scraped data, email alerts)

- Deep research and planning for personal stuff

gptel + gptel-agent in emacs, hooked up to a Claude API token and ChatGPT oauth provided by work, plus some custom gptel additions (tools and tool invocation inspection mainly).

1) Claude code (Max) at work. Claude code (Pro) personal account for side projects. I do still have a ChatGPT subscription. I'm tempted to cancel it, but I like their app. Gemini is bundled somehow into my google subscriptions. I use it occasionally, but I'd be just as happy to not be paying for it.

2) I do use Openrouter for side projects that need an embedded AI. For example, on Halloween I made a talking pumpkin to greet trick or treaters as they came to the porch.

3) Mostly code. I've tried muse, grok, gemini, codex, and pi w/ openrouter, but I always come back to code.

4) no

5) I have a grok bot account I use to manage personal stuff. It organizes my email and calendar, keeps me up-to-date on news, does health tracking, trip planning, and shopping research. As an example, there's a website where I order lunch for my kids. I like to know each morning if I need to pack their lunch, so I have grok bot log into the website and sync it to my calendar. I then get a notification in the morning with everything I need to start the day, like whether to pack lunches, what the weather will be, or if I need to help my kids study for a test.

Tried just about everything over the last couple of years. Settled on the unified Cursor, Grok, Grok Bot plan for about $200 / month. Doing everything within herdr for programming and Grok Bot for all non-programming related AI life stuff.

It's a solid groove. Grok Bot helped me get so much stuff off of my plate.

Deepseek Flash or GLM flash, paid upfront using Openrouter, via Cline on Visual Studio Code.

1) GPT plus, z.ai GLM Coding Lite-Monthly Plan ( legacy ), Opencode Go

2) personally, no. At work there is a Gemini Enterprise and Kiro Pro+ subscriptions

3) pi.dev inside herdr; both at home and work. I just pile the APIkeys and stick to same harness.

4) mostly, pi-subagents has an orchestration mode

5) no.

Herdr, Pi, OpenRouter. I haven't found that I've needed much more than this for quite some time now.

Take a look at gnosh.ai; it is a frontier harness worth understanding. You will hear about it in the coming weeks.

$100 ChatGPT sub $100 Claude sub

Most coding via Opus 5.5.

Codex for text and some non-coding tasks.

Separate virtual desktop and clone on dev drive with VScode for each task (much prefer its diff tool and use it to launch builds etc).

I never come close to running out of tokens because I still do code reviews often enough to throttle usage.

I'm playing around with some ideas on top of pi durable. Cascading contexts exposed over http. Mostly deepseek, glm and kimi in there. https://github.com/resolveworks/key

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

1) None

2) Openrouter. I don't have a preference right now. All the top open-weight models work fine for my projects. Currently using Mistral.

3) I used pi for a while and lately I started using openrouter's ori coding agent. Working great so far.

4) None

5) None. I tried Muse for a while and it was too slow to do anything useful.