All Agents Considered is for people who use AI every day and want more control over the systems they rely on.
I write about open source agents, local models, cheaper provider stacks, memory, automation, and the practical setups that make AI useful without tying everything to one closed subscription.
This is written for people who understand work, even if they do not live in GitHub.
The kind of work where you juggle inputs and make decisions. Where you hand things off and check results. Where you waste an afternoon because a tool forgot the context, hit a limit, or trapped the useful part behind someone else’s interface.
AAC exists to make those setups understandable, useful, and more under your control.
What You’ll Get
Each week, I send practical writing built around real tests and the lessons hiding behind them.
You’ll get:
- articles that show what a setup replaces and what it gives you control over
- open source agent workflows explained step by step
- local model and provider breakdowns with plain-English tradeoffs
- prompts, configs, templates, and setup notes from actual projects
- honest notes on what helped, what broke, and what was worth skipping
- examples of where a human still needs to check the work
- community chat access and direct support from me (soon)
- courses that teach you how to set up and run AI agent systems, starting with Hermes 101 (soon)
The point is to make AI easier to reason about and less dependent on a single company’s interface.
What I Mean by Agents
When I say AI agents, I’m talking about systems that go beyond answering a single prompt. They follow multi-step instructions and work with files. They hold context across sessions and run their own checks. They hand work back to you, sometimes well and sometimes confidently wrong.
The real work happens in the gap between a demo that impresses you and a setup you rely on during a normal week.
AAC focuses on that setup: the memory, files, models, tools, checks, and handoffs around the agent.
Who This Is For
AAC is for people who want AI to be genuinely useful in their daily work. Founders and operators. Writers and researchers. Consultants and solo-business owners. Anyone who moves work from one step to the next and wants the tools to keep up.
You do not need to be an engineer. Some posts get closer to tools and building. Others focus on the habits and checks that make AI more reliable over time.
The shared thread is control: knowing what your AI setup does, what it costs, where it stores context, and when you should step in.
About Me
I’m Dan.
I learn by building things and watching where they help versus where they waste time. I spent years using AI through other people’s interfaces, then started moving more of the useful work into systems I understand and control.
I teach through All Agents Considered because open source agents, local models, and practical workflows deserve clearer explanations than hype threads and vendor docs.
Questions
Reply to any email if you want to ask something, push back, or tell me what you’re building.
I read every response.