# SageOx

> The hivemind for human–agent teams. SageOx adds shared, structured team memory that AI agents consult before they act, so agents don’t drift, humans stay in flow, and the team moves forward together.

## Product
- [Homepage](https://sageox.ai): How SageOx turns humans and AI agents into real teams
- [Product overview](https://sageox.ai/product): Capture decisions seamlessly as the team works; surface context wherever the team needs it
- [About](https://sageox.ai/about): The team building SageOx
- [Demo](https://sageox.ai/demo): See SageOx in action

## People
- [Ajit Banerjee](https://sageox.ai/about/ajit-banerjee): Founder & CEO; writes on agentic engineering, how we work, infrastructure
- [Ryan Snodgrass](https://sageox.ai/about/ryan-snodgrass): Founder & CTO; writes on agentic engineering, how we work, ai literacy
- [Milkana Brace](https://sageox.ai/about/milkana-brace): Founder & CPO; writes on agentic engineering, how we work, book
- [Galex Yen](https://sageox.ai/about/galex-yen): Founding Engineer
- [Emory Clark](https://sageox.ai/about/emory-clark): Founding Designer; writes on agentic engineering, design, how we work
- [Madhur Shrimal](https://sageox.ai/about/madhur-shrimal): Engineer; writes on agentic engineering, announcements, hivemind
- [Faridun Mirzoev](https://sageox.ai/about/faridun-mirzoev): Forward Deployed Engineer; writes on agentic engineering, how we work, team
- [Rupak Majumdar](https://sageox.ai/about/rupak-majumdar): Advisor; writes on agentic engineering, formal methods, research
- [Violet Snodgrass](https://sageox.ai/about/violet-snodgrass): Intern; writes on agentic engineering, how we work, team
- [Alex Helgeson](https://sageox.ai/about/alex-helgeson): Founding Business Ops Manager

## Documentation
- [Docs](https://sageox.ai/docs): Guides, references, and cookbooks for building with SageOx
- [Docs llms.txt](https://sageox.ai/docs/llms.txt): Full docs section index, grouped by topic
- [Docs llms-full.txt](https://sageox.ai/docs/llms-full.txt): Every docs page concatenated as Markdown, for one-shot ingestion

## Changelog
- [Changelog](https://sageox.ai/changelog): Everything SageOx has shipped, newest first, across console, CLI, mobile, and Ox Dot
- [Codex or Claude as the brain](https://sageox.ai/changelog) (2026-09-14): Pick the model behind a coworker, and a handoff between two of them stops echoing.
- [Markdown that renders on Slack](https://sageox.ai/changelog) (2026-09-12): Emphasis, headings, lists and links arrive rendered instead of showing their punctuation — nothing to turn on.
- [A lighter public site, and docs that ship on their own clock](https://sageox.ai/changelog) (2026-09-11): The homepage, blog, changelog and docs keep every URL they had, load lighter, and now ship without waiting for a product release.
- [A scheduled job that can think](https://sageox.ai/changelog) (2026-09-11): A job's body may be a prompt, so a coworker can do light scheduled work only its brain can do.
- [Say it once, and know why it didn't run](https://sageox.ai/changelog) (2026-09-10): A job announces without keeping state, a parked job reads as parked, and a manifest can be checked before it merges.
- [Activity is on for every team](https://sageox.ai/changelog) (2026-09-09): The Activity card on your team home page — who moved what today, in plain sentences, with links back to the source — is no longer limited to an early-access coh
- [A job can hand back an answer](https://sageox.ai/changelog) (2026-09-08): Reviewed jobs return structured results through job_status, and a run can leave a record you can count.
- [Jobs on their own machine, and a ledger it can read](https://sageox.ai/changelog) (2026-09-07): Work runs in its own pinned image, and the coworker can ask what the week held — or say why it can't answer.
- [One clock, and a gate in its own words](https://sageox.ai/changelog) (2026-09-04): The brain's tool calls are bound by the turn, and a passing gate's sentence can post as written.
- [Reads the room, answers where it was asked](https://sageox.ai/changelog) (2026-09-03): It can list its channels, name who's in them, wake one person — and answer in the thread that asked.
- [Sharper on Slack, steadier on schedule](https://sageox.ai/changelog) (2026-09-02): Mentions arrive as names, missed DMs get answered, and a scheduled tick runs wherever the cluster puts it.
- [An AI coworker in your team's channels](https://sageox.ai/changelog) (2026-08-31): Open source and yours to run: one AI coworker across Buzz and Slack, starting from your team's memory.
- [Notifications in the top bar](https://sageox.ai/changelog) (2026-08-25): New discussions, team events, and merge suggestions collect in one place, so you can catch up on what happened while you were away.
- [A clearer account home and a redesigned Security page](https://sageox.ai/changelog) (2026-08-23): Your settings reorganized around what needs attention: expiring credentials, a security posture at a glance, and provenance on every key.
- [Faster, sharper transcripts](https://sageox.ai/changelog) (2026-08-19): Transcripts arrive within minutes of a Discussion and read cleaner, with steadier speaker labels, including live transcription for Discussions captured through 
- [Live Highlights shows where the Discussion has arrived](https://sageox.ai/changelog) (2026-08-17): A fullscreen room surface that shows the through-line of a Discussion as it happens, with recent highlights, who's speaking, and a QR to join.
- [New ways to sign in, and control over your account](https://sageox.ai/changelog) (2026-07-30): Continue with Apple, a verification code at signup, and self-service email changes.
- [Visual plans your team can act on](https://sageox.ai/changelog) (2026-07-28): Your AI coworker authors a visual plan; your team reviews it, and it stays linked to the PR it shaped.
- [Team Vocabulary teaches transcripts your words](https://sageox.ai/changelog) (2026-07-28): Add the names, products, and jargon your team says out loud, and transcripts stop mis-hearing them.
- [Photos and files, right in a Discussion](https://sageox.ai/changelog) (2026-07-22): Add a photo or a file to a Discussion from the camera button when you join, or from the Photos card afterward, and see it inline with the transcript.

## Blog
- [Blog index](https://sageox.ai/blog): Insights on agentic engineering, team context, and building with AI coworkers
- [One week inside SageOx](https://sageox.ai/blog/one-week-inside-sageox): A one-week work trial instead of an interview loop: shipping on day one because the context was already in the system, Friday demos that crossed designer and en
- [From a Whiteboard to NVIDIA](https://sageox.ai/blog/thoughts-on-nvidia-hf): Infrastructure only gets noticed when it breaks. On Onam, eating sadhya with my mother, my phone lit up with the rumour that NVIDIA would acquire Hugging Face. 
- [Introducing the SageOx Agent Toolkit](https://sageox.ai/blog/introducing-sageox-agent-toolkit): We're launching an open-source, Buzz-compatible toolkit that brings SageOx's team memory to your hosted agents — so they arrive with your team's decisions, conv
- [Two weeks of learning at SageOx](https://sageox.ai/blog/two-weeks-of-learning-at-sageox): A rising high school junior spends two weeks at SageOx — auditing the website, sitting in on design meetings, and learning what it actually means to work alongs
- [Reflections on my work trial](https://sageox.ai/blog/reflections-on-my-work-trial): Leaving a job to build with LLMs, a four-day work trial instead of a traditional interview, and shipping a hosted Buzz agent wired into SageOx's team context, t
- [A reminder for the solo designer from Seattle Tech Week](https://sageox.ai/blog/a-reminder-for-the-solo-designer): I'm the only designer on a team of six, and some days the only designer I talk to is Claude Design. Spending an evening in a room full of designers gave a remin
- [Trust at 40x](https://sageox.ai/blog/trust-at-40x): AI makes execution abundant. It does not make a team believe the same outrageous thing at the same time. Trust does that. How radical transparency lets a team m
- [Welcome, Buzz. We've been expecting you.](https://sageox.ai/blog/welcome-buzz): Block launched Buzz — an open workspace where humans and AI agents work side by side as equals, same identity, same permissions, same audit trail. This is the w
- [The New Agentic Order — Chapter 2: The Evolution of Coordination](https://sageox.ai/blog/the-evolution-of-coordination-chapter-2): Every coordination tool a software team has ever used was an answer to a question about scarcity. A genealogy of waterfall, agile, DevOps, and platform engineer
- [What has Claude NOT changed in 2026](https://sageox.ai/blog/what-claude-has-not-changed): Everyone is asking what AI coding has changed. The more useful question (the one Bezos taught me to ask) is what it hasn't. Two stories from one week: a Heisenb
- [MoQ: SageOx's invisible infra](https://sageox.ai/blog/infra-as-unlock-moq-realtime-presence): The most important technology in a product is usually the part the customer never sees, built by someone solving a different problem. Alan Frindell spent years 
- [Avi Agola reflects on his internship at SageOx and shares a walkthrough of his project](https://sageox.ai/blog/avi-agola-internship-reflection): Avi Agola wraps up his internship at SageOx — sharing what he built, what he learned, and what's next.
- [Welcome to SageOx, Alex and Emory!](https://sageox.ai/blog/welcome-alex-emory): We're excited to welcome Emory Clark and Alex Helgeson to the SageOx team.
- [YAGNI changed altitude](https://sageox.ai/blog/yagni-changed-altitude): YAGNI isn't wrong — the floor it operates on has risen. When implementation cost collapses, scoping discipline belongs at the systems level, not the code level.
- [The New Agentic Order — Chapter 1: The New Shape of Scale](https://sageox.ai/blog/the-new-shape-of-scale-chapter-1): Scale is no longer about how many people you have. It's about how much throughput moves across your team relative to the attention you can bring to bear on it. 
- [Oxy goes to art school](https://sageox.ai/blog/oxy-goes-to-art-school): A new idea is the fastest-acting antigen known to science — and the deeper problem is that we can't even perceive an idea that doesn't fit what we already belie
- [RIP tech interviews, Oxy will not miss you](https://sageox.ai/blog/rip-tech-interviews): Every engineer we've invited to a work-trial would have been a confident 'hire' under the methods I trusted my whole career. Only two of six clicked. This is wh
- [The future of software: Steve Yegge and Ajit Banerjee in conversation](https://sageox.ai/blog/the-future-of-software-yegge-banerjee): Steve Yegge and Ajit Banerjee on AI literacy cohorts, the death of job titles, why human connection is getting more important — not less — and what it actually 
- [Scribe API powering SageOx](https://sageox.ai/blog/scribe-api-powering-sageox): Ajit Banerjee and Zoom AI's Zhenbin Xu on how Zoom's brand-new intelligence APIs — topping Hugging Face's ASR leaderboards — are now powering SageOx's real-time
- [What happens to software engineering when generating code is free? (Part 1)](https://sageox.ai/blog/when-generating-code-is-free-part-1): Dr. Rupak Majumdar — Scientific Director at the Max Planck Institute for Software Systems and SageOx Advisor — on the collapsing cost of code generation, the va

## Feeds
- [RSS](https://sageox.ai/rss.xml): Blog updates (RSS 2.0)
- [Atom](https://sageox.ai/feed.xml): Blog updates (Atom)
- [Sitemap](https://sageox.ai/sitemap.xml): Machine-readable URL index
- [Agent catalog](https://sageox.ai/.well-known/ard.json): Agentic Resource Discovery manifest of SageOx's MCP servers, API, and docs for AI agents

## Optional: about SageOx
- SageOx is the hivemind for human–agent teams. It captures discussions, agent sessions, and assistant chats as durable team artifacts, then primes every new AI agent session with the team’s shared memory so agents stay aligned as the project evolves.
- Based in Seattle. Raised $15M seed in May 2026 led by Canaan Partners, with A.Capital, Pioneer Square Labs, and Founders’ Co-op.