I’m tired of these AI written articles man. There could be information worth sharing here but having it all generated makes me assume you don’t care. So neither do I.

I like the idea of using claude code in a separate git-managed repo. I do this for the management side of my job, and use Obsidian as the human interface on top.

However I would never use someone else's starter repo. The whole point is to augment my own perspectives and strengths. On top of that, the AI written pitch is extremely off-putting. If you can't even bother to take the time to write your own pitch, why would I believe for a second that your workflow has anything unique that goes beyond the lowest-common denominator AI capabilities.

One man's starter workspace is another man's context pollution.

There is danger in over-fitting the agent. Fast release cycles require frequent model switching, sometimes to another provider altogether.

In the end, it's hard to measure improvement. Most of the advice out there is vibes based, treat it as such.

I'm working on something similar called devtop. You run it via docker and it has a UI for viewing artifacts and talking to an LLM.

The default workflow is 1) you create or derive documentation from the source code 2) PRD is derived from docs 3) tickets are derived from PRD 4) agents can work on tickets. All artifacts are files which live in the repo.

https://github.com/synlace/devtop

Edit: Still under active development.

Claude.ai Web (non-Code) has this feature where you can write into /areas/ or /topics/ and these files are automatically updated in other chat's sandboxes (but not re-loaded in those chats).

You cannot write the files directly nor can you download them directly, but you can ask a chat to show you the content or only the mermaid diagram that exists in workspace.md (or however you want to name it), and mermaid can represent quite complex networks out of the box, their relations, what is blocking something else, and so on. Chats can then edit this file with search-and-replace instead of re-writing it all over again.

I transitioned into a Product role this year, coming from a CS/SWE/DS background and it is just shocking how poorly the AI models perform at written content.

Any product manager worth their salt should be in a coding interface--the line between Product and Engineering is blurring a lot more now, and to stay competitive PMs will need to reach a little further into the technicalities. Otherwise you risk just being another “idea guy” pushing tickets around, basically customer support on steroids which is getting eaten by Agents. Connecting the ideas to something tangible and translatable that an Agent could not do themselves is the new game.

(I am a PM of 15 years trying to navigate all of this…)

This blog post immediately reads as AI slop. Why bother reading it if they can’t be bothered to write it?

Sometimes I wonder what 'product-level' code really means. What actually is a product?

The code that goes into PLC equipmentss and hardware is incredibly simple—but it's still a product. If you look under the hood of a lot of software, the code is a mess. Still, it's a product. And when Claude's code leaked, everyone laughed at the code hygiene—but a lot of people still used it.

What is a product, really?

why is this on HN lol

[flagged]

[flagged]

[dead]