--- title: My Agent Setup | daily.dev description: A founder shares a personal setup running six specialized AI agents (executive admin, ops, dev, marketing, research, and infrastructure) to help scale multiple... canonical: https://daily.dev/posts/my-agent-setup-8ttve5oux twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: My Agent Setup | daily.dev og:description: A founder shares a personal setup running six specialized AI agents (executive admin, ops, dev, marketing, research, and infrastructure) to help scale multiple... og:url: https://daily.dev/posts/my-agent-setup-8ttve5oux og:image: https://api.daily.dev/og/posts/8TTve5oux.png og:image:alt: My Agent Setup og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # My Agent Setup **[Hacker News](https://daily.dev/sources/hn)** · 8 min read · 1 upvotes · 0 comments ## Summary A founder shares a personal setup running six specialized AI agents (executive admin, ops, dev, marketing, research, and infrastructure) to help scale multiple products and a nonprofit with fewer humans. The agents run on Hermes runtimes, use OpenAI GPT-5.6 models, communicate via the open-source Slack alternative Buzz (built on the Nostr protocol), store memory in Mnemosyne and a shared Obsidian wiki, and run on a single DigitalOcean droplet secured with Tailscale. Concrete use cases include automated Sentry triage, morning briefings, social calendar review, and a daily work report summarizing all agent activity. The author admits the setup took roughly 10x longer to build than doing the underlying tasks manually, but frames it as a long-term investment. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### Why would someone switch from Anthropic Claude to OpenAI GPT models for running autonomous agents? A subscription-based OpenAI plan (starting at $100/month, with a $200 tier available) supports the always-on agent usage pattern, whereas Anthropic's Fable model does not allow subscription usage for this kind of setup, leading to unexpectedly large API bills when used that way instead. Frontier models are largely interchangeable for this type of agent work, making cost and billing terms the deciding factor. _Comparing model pricing and subscription terms for always-on agents is exactly the kind of decision daily.dev helps developers track._ ### What is Buzz and how does it relate to the Nostr protocol? Buzz is an open-source Slack alternative built by Block with first-class support for AI agents as team members, currently at version 0.5.9. It runs on the Nostr protocol, where NIP-01 defines a universal signed event format, key-based authorship, WebSocket relays, and filterable real-time subscriptions; agents in Buzz are represented simply as keypairs, and any community can host its own relay. _Developers evaluating open, agent-friendly chat protocols like Nostr can follow daily.dev for tools built on it._ ### What hardware is enough to run six always-on AI agents continuously? A single DigitalOcean Basic Droplet with 4 vCPUs, 8 GB RAM, and 160 GB disk was sufficient to run six specialized agents continuously, secured with Tailscale and no exposed public ports. An earlier attempt with only 4 GB of RAM caused the box to swap to disk, requiring the upgrade to 8 GB. _Sizing infrastructure for always-on agent fleets is a practical concern daily.dev readers researching agent deployments can track._ ## Community take How the wider developer community reacted, aggregated from 1 discussion and 43 comments across hackernews (as of 2026-09-14). **TL;DR:** Reaction is a mix of genuine curiosity about the technical details (costs, context windows, MCP setup) and skepticism about whether such elaborate multi-agent setups deliver real ROI beyond being a fun experiment. **Sentiment:** 30% positive · 45% mixed · 25% skeptical **The case for** - The transparency and lack of hype in describing the setup is appreciated. - Several commenters are running similar personal agent/automation setups and find the domain-separated, least-privilege agent design sensible. - Watching a real-time log-triage/auto-fix agent in action is seen as one of the most compelling use cases. **The pushback** - Multiple commenters agree the personal ROI on these elaborate setups isn't really there yet compared to just using coding assistants directly. - Some argue heavy MCP/tool usage can bloat the context window if not managed carefully. - One commenter frames the broader trend as exhausting over-engineering ('bikeshedding maximization machine') rather than genuine progress. - A stray accusation that the post is an ad, though unsubstantiated. **By community** - hackernews (mixed): Discussion splits between technical enthusiasts comparing notes on MCPs, context windows, and costs, and skeptics questioning whether this kind of setup is worth the effort. **Hottest debate:** Whether elaborate personal multi-agent setups like this actually deliver ROI or are mostly an elaborate, fun form of over-engineering. **Open questions** - How well do these setups scale in context-window usage as more MCPs/tools are added? - Is a single executive agent architecture preferable to many specialized agents? - What's the real cost-efficiency tradeoff versus cheaper VPS alternatives? **Highlights** > Not pointed at the author, but at the current state of affairs: this is fucking exhausting. We went and made a trillions-dollar market out of the bikeshedding maximization machine. > — [moribvndvs on hackernews · 3 comments](https://news.ycombinator.com/item?id=49273099) > Very cool setup. I went down a similar path and while it is definitely cool to see the extent of how agentic AI can be, I agree the ROI isn't _quite_ there yet. I don't live in a world where I need to be constantly reading and replying to emails, so 95% of my inbox is just subscription spam. With development I need to be at the helm to design the planning requirements and actively make decisions before the agent goes off and executes the plan. But I don't need a personal agent for that, I work directly out of codex/claude-code. > — [orangebread on hackernews · 1 comments](https://news.ycombinator.com/item?id=49272873) > The most interesting part of this to me is having an agent that reads logs and spins up fixes in real time. Great write up! > — [owencmcgrath on hackernews](https://news.ycombinator.com/item?id=49274243) > Couple of notes on this First, it matters how well the MCP is written; a well-written one won't be so massive Secondly, different sessions should load different MCPs. Load only what is needed Thirdly, if you're finding you need a dozen MCPs, you need to consolidate that into a single service that does the work and then presents a single, unified API (via MCP) to your agent Finally, start using models with 1M context window limits. I really don't know how people can stand being stuck at a limit like 272K. > — [trollbridge on hackernews · 1 comments](https://news.ycombinator.com/item?id=49274749) **Source threads** - [hackernews](https://news.ycombinator.com/item?id=49272484) · 68 points · 43 comments ## Similar posts on daily.dev - [One Telegram contact to rule my team of AI agents](https://daily.dev/posts/one-telegram-contact-to-rule-my-team-of-ai-agents-qgiolsxxe) · RUBYLAND · 1 upvotes · 0 comments - [OpenAI debuts always-on agents to end the friction of manual team handoffs](https://daily.dev/posts/openai-debuts-always-on-agents-to-end-the-friction-of-manual-team-handoffs-cliyqnsds) · The New Stack · 0 upvotes · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents) [View this post on daily.dev](https://daily.dev/posts/my-agent-setup-8ttve5oux) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"My Agent Setup","url":"https://daily.dev/posts/my-agent-setup-8ttve5oux","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/my-agent-setup-8ttve5oux"},"datePublished":"2026-08-12T14:49:20.434Z","dateModified":"2026-09-14T09:08:44.133Z","description":"A founder shares a personal setup running six specialized AI agents (executive admin, ops, dev, marketing, research, and infrastructure) to help scale multiple...","image":"https://media.daily.dev/image/upload/s--qPvKM23u--/f_auto/v1722860399/public/Placeholder%2009","thumbnailUrl":"https://media.daily.dev/image/upload/s--qPvKM23u--/f_auto/v1722860399/public/Placeholder%2009","isAccessibleForFree":true,"articleSection":"Hacker News","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Hacker News","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/hn","url":"https://daily.dev/sources/hn"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/my-agent-setup-8ttve5oux","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents","timeRequired":"PT8M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Hacker News","item":"https://daily.dev/sources/hn"},{"@type":"ListItem","position":3,"name":"My Agent Setup"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/my-agent-setup-8ttve5oux#faq","mainEntity":[{"@type":"Question","name":"Why would someone switch from Anthropic Claude to OpenAI GPT models for running autonomous agents?","acceptedAnswer":{"@type":"Answer","text":"A subscription-based OpenAI plan (starting at $100/month, with a $200 tier available) supports the always-on agent usage pattern, whereas Anthropic's Fable model does not allow subscription usage for this kind of setup, leading to unexpectedly large API bills when used that way instead. Frontier models are largely interchangeable for this type of agent work, making cost and billing terms the deciding factor. Comparing model pricing and subscription terms for always-on agents is exactly the kind of decision daily.dev helps developers track."}},{"@type":"Question","name":"What is Buzz and how does it relate to the Nostr protocol?","acceptedAnswer":{"@type":"Answer","text":"Buzz is an open-source Slack alternative built by Block with first-class support for AI agents as team members, currently at version 0.5.9. It runs on the Nostr protocol, where NIP-01 defines a universal signed event format, key-based authorship, WebSocket relays, and filterable real-time subscriptions; agents in Buzz are represented simply as keypairs, and any community can host its own relay. Developers evaluating open, agent-friendly chat protocols like Nostr can follow daily.dev for tools built on it."}},{"@type":"Question","name":"What hardware is enough to run six always-on AI agents continuously?","acceptedAnswer":{"@type":"Answer","text":"A single DigitalOcean Basic Droplet with 4 vCPUs, 8 GB RAM, and 160 GB disk was sufficient to run six specialized agents continuously, secured with Tailscale and no exposed public ports. An earlier attempt with only 4 GB of RAM caused the box to swap to disk, requiring the upgrade to 8 GB. Sizing infrastructure for always-on agent fleets is a practical concern daily.dev readers researching agent deployments can track."}}]} ```