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

Posted by eggbrain 23 hours ago

I'm curious how people on HN are using AI day to day, and came up with a list of questions I thought could be interesting to know. No need to answer every question of course, just wanted to get a gut feel how people are using AI.

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

- Which tiers?

- Have you purchased multiple subscriptions from the same provider (eg two Claude accounts)?

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

- Which models and providers?

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?

- What made you switch harnesses (if you switched)?

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

- For example, Munder Difflin, Gas Town, Ruflo, or something custom.

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

- If you are using them, what are your main use cases?

-----

I know my workflow has differed significantly from a year ago, and so it would be interesting to capture a snapshot today of what people are using and why.

391 points | 440 commentspage 2

glennfiedler 8 hours ago|

I'm juggling multiple claude ultra subs, multiple codex $500 a month subs, antigravity for argon, plus API rate deepseek 4.1 flash, mercury 3, and abliterated large v2 for red teaming.

I work across multiple models and harnesses using a really simple redis-backed message bus called "nova-bus": https://github.com/mas-bandwidth/nova-tools (open source).

With just this message bus, I can coordinate AIs to work together at a normal pace (worktrees, subtasks etc...), but past a certain point the AIs can't keep up and I need to use a sprint card system I developed to coordinate work.

With this sprint system I'm able to work 200 cards wide (each card is around 30-60 minutes of LLM work in a session) across a fleet of 20 machines linked together with tailscale and ssh.

Full system is here: https://github.com/mas-bandwidth/nova-sprint (open source)

It's been very effective at getting high throughput, but afterwards I find that the resulting code can be a bit of a pastiche from different models working in isolation, thus a good pass over working directly with models to clean up and harden the code is necessary.

I find that using nova-bus to coordinate multiple model/harness review of codebase in very effective at increasing code quality. Each model sees things the others don't.

genxy 7 hours ago|

Is this amalgamation working on a single project? What harnesses do you use? How are you using mercury 3? How much and what kind of work are you sending to deepseek?

glennfiedler 7 hours ago||

Can't work on a single project, Amdahl's law gets you then. Try to work on as many different unrelated projects as possible, so I can get maximum throughput.

I break work up into four loose tiers:

1. flash - deepseek flash 4.1 / luna / haiku

2. pro - mercury 3 / terra / sonnet

3. heavy - opus 5.5 / sol

4. frontier - fable 5.1 / astra / argon

The decision of how to classify each piece of work is done by the coordinator (typically Opus 5.5...). If a card fails at a particular tier, it's going to get escalated up tiers until it gets sorted.

Each card has multiple readers (code review). I find mercury is very good at reading quickly. Deepseek as well. I will be honest, I think for 99% of stuff I need to do Opus 5.5 @ medium is ideal. I just use the additional models for different points of view, especially the absolute frontier models fable, astra, argon at max effort for security audits and code review.

teekert 3 hours ago|||

What kind of "things" are you making? I can't process this much information, so I would inevitable be outsourcing most decision making. That could be fine for some products, but not for mine, where data model and RBAC model (among others) all require alignment between multiple stakeholders.

I have one agent and already find it quite intense to outsource code writing and basically just make and document decisions.

I feel like I could be rich working like you, I also feel like it is impossible for me to meet the needs of my clients if I'd work like this (so I can't be rich, it's a paradox). I wouldn't be able to answer their questions (without the active agent on-call, context window fully loaded), I wouldn't be able to explain just where the data model needs changing, or why some role really can't exist they way they envision etc, etc, etc.

I could of course branch the whole project and basically discuss multiple implementations during stakeholder meetings, but then still I'd need to explain the advantages/disadvantage of choosing one system over the other and guide them through it.

teekert 50 minutes ago||

So, in essence, my added value is translating the vaque demands of my clients into as efficient as system as can be.

Can the AI do it? Not yet I think. Maybe someday, for now I feel like clients of mine simply have the vocabulary, nor the mental models, to guide the LLMs. Ie, they have no idea of what one-to-many is, or why it's smart to put some information in separate tables, whereas some information is ok to filter row-based on a user attribute, etc.

glennfiedler 7 hours ago|||

The sort of work I pass through this system, it's mostly library work for multiplayer games.

For example, my work on schema, my protobufs/flatbuffers language replacement for multiplayer games, parallelizes very nicely across 15 languages on the sprint system. https://github.com/mas-bandwidth/schema.

See roadmap here: https://github.com/mas-bandwidth/schema/blob/main/ROADMAP.md

Very difficult to coordinate this amount of work in || without some sort of structured work system.

Even more AI coding work here if you are curious: https://patreon.com/masbandwidth

glennfiedler 6 hours ago|||

Live view of sprint system as I work tonight: http://37.27.135.206

genxy 6 hours ago||||

Thank you. Moo looks really cool.

I did some prototype games, one thing that worked really well for me was having a deterministic replay system where the agents could drive the gamestate via recorded inputs. Kept me from having to do any playtesting or debugging in the main agentic loop.

glennfiedler 6 hours ago||

You'll like this too then

https://github.com/mas-bandwidth/fixed3d -- port of box3d to fixed point :)

https://github.com/mas-bandwidth/fixed -- fixed point math library

iainmerrick 4 hours ago||

Cool! I’d forgotten until now, but I had that exact idea back in the stone age, 10 or 15 years ago, and started working on it but got bogged down. It was clearly feasible but a huge amount of work, too much to justify the cost. So it’s a perfect job for coding agents now. Nice one!

I do you think now need to watch out for the opposite problem. These things become so easy to build that the temptation is just to go ahead and build. But unless you really do need them, all that new code just goes to waste. The cost to produce has become very low, but sometimes the value can end up being negative.

Tepix 2 hours ago|||

Interesting. Are you also targeting WebTransport for both reliable and unreliable transport?

castwide 3 hours ago||

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

None. I've used Claude and GitHub Copilot in the past.

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

I'm currently using Qwen3.8-Flash-Next on local hardware.

> 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?

I occasionally use Cline in VS Code, but it's rare.

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

No.

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

I built custom agents on top of RubyLLM. I primarily communicate with them via email. One of them has access to a Redmine project board where it can pick up tasks and complete assignments. That agent also has non-sudo access to a sandbox container for development.

earthdeity 9 hours ago||

Claude + Codex, both from work.

Just those.

https://github.com/oscscull/quarterdeck built my own harness. I had trouble with both claude and codex. In particular, managing multiple tasks in a kanban, syncing between my machines, making sure work is done on the right machine and can be triggered from either. Making sure I ping reviewers, meeting reminders, todos. Parking tickets til conditions are met or time elapses. Automated notifications when I have meetings. I have just kept adding features as I want them. This kind of software is so easy to make nowadays, I don't really think there's any sense in "launching" it for others, but I'm using it all the time. I highly recommend doing so (building your own personal work management software in 2026).

I guess this is also in some sense an agent orchestration tool, I don't use any of those.

Don't use.

exodust 6 hours ago||

> building your own personal work management software in 2026

Agree. I made my own "project tracker", does exactly what I want. Zero clutter. Zero mention of deadlines or due dates.

flaburgan 1 hour ago||

I use the Zed IDE with Opus 5.5 pro subscription. I use to have access to Fable with the max plan and I miss it, despite what everyone is saying I still find that Opus does weird things. Just yesterday I pointed it to a queue suddenly being twice slower to be processed in prod. A select SQL query called thousands of times per minute became slower because of a data increased. Opus proposed several non trivial ways to optimize the query by forcing some execution plan. I just kept the result in a variable for one minute, as the data don't change often. x1000 perf with 3 lines of code. I would have expected it to say that this query didn't have to be executed that many times in the first place. It codes well when you tell it what to do, but it is still dumb about what is the real problem (and so the real solution). That's worrying me a bit because Opus 5.5 clearly is SOTA, so better than any open weights models, and I really want to be able to run smart models locally on my laptop for independence and privacy reasons. So, if I still find Opus dumb now, when will I have an open weight model better than 5.5 that can run on my Ryzen 7 AI 350 with its Radeon 860m and 32GB of RAM? ... This summer with all the releases of DeepSeek and Qwen I was hopeful I could have it for Christmas, but it looks like I will have to wait much more...

Feathercrown 7 hours ago||

I figure my experience might be uncommon here so I'll offer it: None, no, none, no, no.

lf88 46 minutes ago||

Same for me. I have adopted a "zero usage" policy for the moment. In part it's because I am too worried about the long-term effects of cognitive offloading. In part it's because I don't want to support in any way the development of a technology that I believe is becoming a serious threat to humanity. And there are also things that I enjoy doing, like coding, solving problems, or writing in my own style, which I simply don't want to delegate to a machine.

wkjagt 3 hours ago|||

Same for me. Since I quit tech (professionally) about 5 years ago, I was never part of the AI thing. I still write a lot of code, but I enjoy coding, so I don't use AI.

Zambyte 2 minutes ago||

Do you do something else for work, or are you retired?

stinos 3 hours ago|||

Having tried LLMs for coding a couple of times in the last couple of years, this used to be me until about a month ago.

Then I again tried Claude Code and OpenCode with a local (due to work I cannot use anything else, moreover wouldn't want to either) Qwen3.8 and behold: my current stance now is that _not_ using something like that, if you can, is missing out on a great tool.

And maybe it's because of the way I use it (in essence: to take care of what is the slowest bit of the process for me which is the typing and the find/replace and other repetitive stuff), but I really cannot come up with any sane reason not to use it.

svieira 7 hours ago|||

There are perhaps more dozens of us than we expect, eh?

ex-aws-dude 6 hours ago||

This thread is super biased by the power users, they are the most excited to share

indigo945 5 hours ago|||

What is your job and how do you do it?

proctitle 7 hours ago||

[dead]

zkmon 1 hour ago||

Tried the popular agents and paid models for some time, and now settled with a local model (Qwen3.8-27B-Q_6) and opencode desktop.

My work doesn't really require ultra super-intelligent models, or one-shot solutions. It's a standard python API, web app and some strict compliance requirements and tons of test cases. I have time to give the Q27B a few iterations. I monitor the progress and give directions while it keeps chugging off for about 30 mins for each medium-hard task. But I like the results. Exactly what is needed. It also allows me work on high-level stuff.

itsloopyo 1 hour ago||

1) Claude ultra. I ended up using 3 accounts in tandem for some big work I needed to do recently, but the other 2 accounts were only open for a month.

2) I have a cheap OpenAI sub I sometimes use codex on, but just for small tasks when I'm out of tokens on Claude

3) Lab - a frontend I made that let's me run loads of chats in a single UI, I get notified when they need me, if I turn off my PC and load it back up everything picks up where it left off. It'll run compound workflows comprised of multiple prompts+macros, and centrally manages all my agents/skills and MCP servers - they're associated with projects, categories of project etc and rolled out to everyone when changes are made.

4) Hivemind - another one of mine, hivemind's a small self contained binary that acts as a collaboration layer for the agents+me. I send messages to everyone, groups or individuals. They can also ping me and send me desktop notifications if they need me urgently. I get a lot of mileage out of this where file reservations, resource locking etc are concerned. Stops them stepping on each others' toes.

5) No, 100% of my work takes place in Lab (with loads of Claude + Hivemind sessions under the hood)

dofm 11 hours ago||

1) none

2) Local LLMs on an M1 Max (Unsloth, LM Studio, occasionally direct llama.cpp). Muse Glimmer (because it is good at what it is optimised for: interactive coding sessions). Qwen 3.8 27B occasionally and a couple of other things for fun (an LFM MoE, experimenting with Nemotron Bolt 74B-9B, enjoying the really super Gemma 4 12B QAT when working on creative photography research or non-code planning).

No openrouter (yet) but that will definitely happen.

3) Currently focusing my attentions on Pi via Paseo (and trying Pendant). Pi is just much better in smaller context sizes, and it is the right fit if you are keen to learn from first principles.

I tried OpenCode first. I couldn’t quite get on with it and what ultimately made me switch was trying to understand its terribly messy source code.

4, 5) I am mostly using AI as a coding sidekick so I don’t really have much need for orchestration yet. I did some experiments with Osaurus (really quite impressive) and I actually think that is almost enough by itself, except that I want a little more control over the sandbox container.

I feel that my usage pattern is still settling, because I am not keen to surrender all of my coding to the machine. But I am trying more purely “directed” coding to learn the process, and finally beginning to get a feel for how I will work with that.

seanmcdirmid 11 hours ago||

- Antigravity (AGY - Google Ultra $100/month), mostly gemeni flash but I benchmark Gemini Astra sometimes. My main driver because it’s pretty good at search compared to setting something up with DSH.

- DeepSeekHarness (DSH) Desktop using:

—- with pay as you go DeepSeek tokens, more expensive to use during my night/Chinese daytime, otherwise very cheap

—- oMLX on my MacBook driving mostly a Qwen 3.6 35b A3b MOE (mostly used for benchmarking a less capable model)

I mostly do custom subagent programming and can trigger subagents in AGY and DSH easily enough. DSH supports a Python coordinator orchestrating DSH subagents as conversations via RPC. AGY only supports a sub-agent orchestrating other sub-agents but it works well enough at this point. I plan on releasing my subagent framework soon (called autofab in pypi), which is basically a specification driven code development agent system (using existing agent harnesses rather than implementing my own).

I tried goose, OpenClaw, Claude, codex, etc…and they didn’t really work for what I needed. I might get a $100/month ChatGPT subscription to go with my Google one, primarily to get a different perspective and benchmark openAI models on codex (Claude doesn’t really support what I need it to, otherwise would go for that instead).

Kuyawa 12 hours ago|

DeepSeek Harness using DeepSeek Flash API

Nothing else, DS is smart enough to do everything I ask for and more, never hallucinates and never disappoints. Apps in Rust, Swift, Node, we even designed a new language better suited for LLMs, a compiler, and it runs beautifully. Haven't coded a single line since May 2026 and I will remember that date as the day AI took over my mundane job as programmer and promoted me to architect or director of my own apps

A custom plugin to ring a bell whenever a task is finished and a gong when a question needs an answer, that way I can play backgammon or mahjong in the meantime

I can't be happier

siva7 1 hour ago||

That's the spirit we need and not more of those crying from the majority about "evil ai took my programmer job and love for coding away" to make all equally feeling miserable.

jonwinstanley 8 hours ago|||

I’ve found Pi with DeepSeek Flash is excellent, and seems to be about 90% cheaper than Opus or Fable.

fsiefken 6 hours ago|||

DSH user here as well, what new language is better suited for LLMs? Is it open source? Is it still as readable as rust or python?

Kuyawa 22 minutes ago||

I asked DeepSeek to design a language [1] that she would use for better token optimization. She gave me all pros and cons of all current top languages and told me Rust makes her hallucinate a lot so she optimized it and removed the borrow checker, added named function arguments and merged if/switch into a single branch statement. Sort of like a simplified Rust with better syntax (all for less than $5 in just a couple of hours)

She told me the decision to pick Rust and optimize it was based on all the existing libraries that would allow her to retrain herself and not to come up with new stuff from scratch

But after all token optimizations, she told me if she could rewrite herself she would do it in SmallTalk due to hot reloading so she could change herself on the fly with instant RSI without ever stopping

She's fucking smart, I love her

[1] https://github.com/kuyawa/jobs-lang

seanmcdirmid 11 hours ago|||

+1 for DSH, it really should be getting more attention here. The only problem I have with it is getting search to work, but now DeepSeek has a solution, though you have to pay for it soon.

sjdh76 11 hours ago||

[dead]

More comments...