Show HN: SageOx – The Hivemind for Agentic Engineering(sageox.ai)

Show HN: SageOx – The Hivemind for Agentic Engineering

https://sageox.ai/blog/introducing-sageox

We built SageOx to solve the alignment problem we kept hitting while working with coding agents.

Claude can generate code quickly — but they don’t share team memory.

Each session starts from scratch. Architectural decisions made yesterday aren’t visible today. A technical debate disappears unless someone manually documents it.

Speed increases, but drift leads to architectural entropy and compounding rework.

What SageOx Does

SageOx provides shared, queryable team memory that humans and agents automatically draw from before they act.

Capture We capture intent as it emerges, always with permission: - Technical meetings - Product discussions - Human–agent coding sessions

Structure Architectural decisions, constraints, conventions, and implementation reasoning become durable, searchable artifacts.

For example, if two engineers decide to standardize on git-lfs instead of git for our media artifacts, that decision (and its rationale) becomes searchable context for future sessions.

If a developer collaborates with Claude to implement a feature, the reasoning behind the implementation becomes part of team memory — without anyone manually writing documentation.

Consult When you start Claude, ox gets primed and automatically retrieves relevant team context — recent decisions, architectural constraints, related discussions — and injects it into the session.

There’s also a web app for reviewing structured context, managing members, connecting repositories, and inspecting the ledger.

Building in Public (Open Work) - come check us out

Demo: https://sageox.ai/blog/introducing-sageox

The Ox CLI itself is built using SageOx, and signed-in users can see: - The debates behind technical decisions - Trade-offs we considered - Moments where we changed direction - The human–agent sessions that produced specific changes

Not just what we shipped — but how we reasoned our way there.

We think this level of inspectable reasoning becomes important as more engineering work is done through AI agents.

Try it! Right now, SageOx v0.1 is for Claude users building entirely through prompts.

If you’re coordinating across engineers using coding agents and seeing drift or repeated decisions, we’d appreciate feedback.

>_ Claude Prompt: Take a look at gh sageox/ox and install the cli

>_ Claude Prompt: ox login

Happy to answer technical questions about architecture, context capture, retrieval, or tradeoffs vs. traditional documentation.

[email protected] sageox.ai https://github.com/sageox/ox

3 comments

port80805 ay önce

Founder here.

Our bet is simple: - In the future, new projects won’t just start with git init. - They’ll start with git init and ox init.

Code gives you versioned artifacts.You’ll also need versioned context.

SageOx is our attempt to make shared, inspectable team memory a first-class primitive — something agents automatically consult before they act.

Happy to go deep on architecture, tradeoffs, or why this might be wrong.

cranberryturkey5 ay önce

Interesting approach. The session amnesia problem is real — every time you start a new Claude session, you're essentially onboarding a new team member who knows nothing about your project.

Question: how do you handle conflicting decisions? If engineer A decides on approach X in a session on Monday, and engineer B decides on approach Y on Tuesday, does SageOx surface the conflict or just store both as valid context?

Also curious about the retrieval quality. The hardest part isn't capturing decisions — it's retrieving the RIGHT context at the RIGHT time. Too much context and you're back to the LLM ignoring half of it. Too little and you miss critical constraints. What's your chunking/relevance strategy?

port80805 ay önce

So far, we have been working on tiny high velocity teams which have immense alignment because we are all colocated.

Many of the issues of conflicts are basically dealt with us hammering them out in our daily standups. We see these as 'merge walls'. Both Ryan and I have been principal engineers in Amazon in our past lives: every day with Claude feels like dropping into a design meeting on a team with 100 engineers. So, we just accept that we have to spend a lot of time whiteboarding and talking as we align our mental models at a daily cadence.

Really good point on the RIGHT context at the RIGHT time. With the first release, we have focused on the plumbing to collect all the sessions and the transcripts. As we get more data, we are going to step up into the next level of analysis where we collate all the data into the right insight at the right time.