2 ms·

Ask HN: How do you build internal agents at work?

1 points by thekafkaf 9mo ago

I’ve been building more and more small agents for different tasks/roles/flows. My recent pattern: start a repo with some prompts + skills, run Codex/Claude Code, then gradually add memory and evals. That usually turns into an iterative loop of improving context, prompts, and tools based on eval results. Curious what others are using: - Any frameworks or patterns that have worked especially well? - Anything that’s friendly for non-technical users, even without a dedicated UI?

verdverm 9mo ago
I use ADK-go (also available in py, ts, java) https://google.github.io/adk-docs/get-started/about/ https://google.github.io/adk-docs/get-started/about/
- thekafkaf 9mo ago
  Did you compare to other SDKs from OpenAI, Anthropic or OSS? I find the harnessing of Claude Code/OpenCode really good, and was afraid of needing to implement a lot of it from scratch if using an sdk, what's your experience in this regard?
  - verdverm 9mo ago
    ADK is lower level then an SDK, it all runs where you want. Using this has let me get at the lower layers of the stack, so I get deeper customization. It has been a great learning foundation. There are various hooks and interfaces, but I had to rebuild much of the higher level UI/UX and such
alexhans 9mo ago
It really depends on the type of agent and problem but your path resonates as a way to slowly build more complexity as you start understanding the uncertain space. I've written about it in my blog with the intent of helping non tech users [1] and also created a site with cookbooks and eval tool comparisons [2]. I've successfully managed to get non technical users into the terminal, even before Agent Skills were a thing. It's absolutely doable. If any of what I share helps or you have any feedback/questions let me know. I'm quite interested to see if the framing works. - [1] https://alexhans.github.io/posts/series/evals/building-agent-skills-incrementally.html https://alexhans.github.io/posts/series/evals/building-agent... - [2] https://ai-evals.io/ https://ai-evals.io/