Over the last 90 days, we captured traces from our coding agents to do continual learning. Here’s what I shared during my keynote on the Continual Learning Loop at the AGI Summit. We plugged Hivemind into every coding agent at the company: Claude Code, Codex, Cursor, OpenClaw, and Hermes. - 93% of tokens are used for computer interaction - 77% of sessions contain at least one correction - 60% of sessions are judged by an LLM to have been resolved correctly More interestingly, 1 in 10 sessions leaked a secret credential to an LLM provider. In-context learning is considered one of the best continual learning methods so far. But how good is it? Apparently, it is 80% as good as weight updates without all the heavy lifting of fine-tuning. Naive SFT on coding-agent traces causes the model to collapse. At best, it delivers only marginal improvements. As coding agents use frontier models, distillation can work into smaller models. But learning from your own LLM traces is very hard. You can easily - misalign the model - forget knowledge - lose capability. Doing this continually is even harder. You can easily run into cumulative catastrophic forgetting. Furthermore, learning a new capability from your own traces does not work. So how can we expect the model to develop new capabilities? How can we dream? One recipe that worked for us was generating the missing experiences based on failures observed in the traces. We then used RL environments to train the model and ran it in shadow mode for three weeks. It outperformed Claude Code 60% of the time. Many people asked, Where should we start? Start capturing your traces now.

Hivemind just crossed 250 stars on github 2K weekly downloads on NPM. 🚀 Connect coding agents to a shared brain > Collect traces into deeplake > Auto-optimize skills > Share across agents, machines and teammates Your agents continuously learn from each other's experience. Get them to compound your intelligence.

Coding agents that actually get better the more your team uses them. Introducing Hivemind: continual learning for AI coding agents. Hivemind turns the traces from every agent your team runs (Claude Code, Codex, Cursor, Hermes, OpenClaw, Pi) into reusable skills, then pushes those skills across all of them. All on your cloud storage. Now with SkillOpt built in, your skills get trained: +19.1 points of accuracy in Claude Code, +24.8 in Codex, best or tied on all 52 setups tested. Open source, one line install.

Agentic LLM systems this year have mostly gone to coding, browsing, research. With the @activeloop (Deeplake) team, we pointed agents at a job they don't usually do: curating a robotics dataset. Roboscribe-AF automates the work a human curator would do — segment episodes, cross-check video vs trajectory, flag the messy ones. Disagreements between reasoners get written back as queryable Deeplake fields. Open source: deeplake.ai/blog/agentfield

Hivemind just crossed 100 stars on github 🌟 github.com/activeloopai/hive… Beyond memory. Let's compound intelligence!

today, we're going beyond memory. your org's agents shouldn't just remember what happened. they should learn from their experience. Hivemind takes agent traces and codifies them into skills every agent on your team can use. > no more explanations > no more duplicate work > no more repeat bugs we make your intelligence compound. here's how 👇

Engineering managers shouldn’t have to play detective every morning. But too often, that’s the job: standups get skipped tickets stay stale and status updates become a scavenger hunt. What if your tools just told you what changed? An AI layer across Claude Code, Codex, and OpenClaw could surface progress, blockers, and momentum automatically. No nagging required. You get your mornings back. Engineers feel less monitored. Everyone stays in sync.