@LangChain@dosu_ai@trychroma@BraceSproul@devstein64@jeffreyhuber On the engineering side, we've already set up a system to automatically update the README on our repos with mermaid diagrams and architecture based on latest commits. I think something like this should exist for customer context, but a repo doesn't feel like the right substrate.
key product decisions are also good sources of context. You could connect our wiki agent or whatever with the GitHub MCP to search for relevant PRs for the customer in question, but that seems a but clunky.
These are the typical places where bugs, feature requests, and nuances about the customer's k8 cluster and deployment are discussed. But stepping back a bit, GItHub PRs and Claude sessions where our engineers and PMs are investigating customer issues, drafting specs, and making
Don't necessarily agree with the statement about 1 year managing 10 agents is better, but def agree it's a great market to be an AI-native player/coach. Can't imagine not being able to do the thing yourself and trying to coach a team on doing that thing.
re: hiring right now
it's a huge bull market for AI-native IC's/player-coaches
it's a huge bear market for "heads of X" managers
never seen such furious bifurcation. to oversimplify: 1 year experience managing 10 agents > 10 years experience managing 10-100 people
πΎ Gemini in @Google Sheets has @claudeai Code-like planning and to-do lists. Behind the scenes I think Gemini is just writing a bunch of #googleappsscripts to do data transformations and cleaning.
Al Chen is on the field engineering team at Galileo. He's not an engineer.
The problem: their product is super technical and their customers ask super technical questions. Docs give the high-level answer, but his customers want the step-by-step answer of how it will work for *their* system.
@bigal123's solution: clone all 15 repos locally, open them in VS Code, let Claude Code answer any question that comes his way.
If you're customer-facing in a highly technical field, this ep is for you.
We also debate the merits of putting @claudeai on a spiff.
As always, ty ty ty to our amazing sponsors
π @orkesio - The enterprise platform for reliable applications and agentic workflows: https://t.co/bhuNNaWXJt
π§ @tines_hq - Start building intelligent workflows today: https://t.co/0U8gh104fA
Watch the full ep on YT π https://t.co/pW1dNmibmO
πAgent Control from @rungalileo is an open source project providing policies and guardrails on your agents down to individual function signatures in your application code (just requires a @control decorator).