Shared memory, seamless collaboration, and sessions that get faster and smarter from first prompt to production
SAN FRANCISCO--(BUSINESS WIRE)--#AI--Kurrent, the company behind KurrentDB, the leading event-native database, today announced the private preview launch of Kurrent Capacitor, an AI-native shared memory solution for coding agents. Capacitor captures every coding agent session — every hypothesis tested, every decision made, every line of reasoning behind every code submission — and makes that record searchable, shareable, and actionable across the entire team and across future agent runs.
Capacitor is free to use and available immediately for Claude Code, Codex and Cursor via a private preview program. Teams can request access at https://capacitor.kurrent.io/.
Agentic Coding is Powerful. Capacitor Makes it Unstoppable
Coding agents — Claude Code, Codex, Cursor, and others — are writing software at a scale and speed that would have seemed implausible even a year ago. The developer's role has fundamentally changed: less author, more director. Prompting, evaluating, judging and coordinating. The productivity gains are historic. And the biggest gains are still ahead.
Agentic coding has already raised the ceiling on what engineering teams can accomplish — and Capacitor raises it further. The opportunity lies in everything that happens around the code: preserving context across sessions, carrying reasoning through handoffs, giving code reviews the grounding to be fast, confident, and accurate. The technology is capable. What's been missing is the shared foundation for humans and agents to fully build together at speed, at scale and with quality.
Capacitor is that foundation. By giving humans and agents a shared memory — recording every event, trace and session, surfacing it through a real-time dashboard, a CLI, and MCP tools that agents themselves can query — Capacitor makes seamless collaboration between humans and agents possible for the first time. Every session builds on the last. Context is never lost. Teams that start using Capacitor today will ship faster tomorrow than they do right now, and faster still the session after that with the assurance of quality and the knowledge that the system is learning along the way.
“Agents have made software development much faster with significant impact while leaving the team’s coordination layer entirely untouched,” said Kirk Dunn, CEO at Kurrent. “Capacitor is our answer to that. It provides interaction between teams of agents and humans in real time. Collaborative, persistent, searchable and secure session memory from the first prompt to production.”
Enable Coding Agents Like Never Before
Capacitor records every coding agent session automatically. No separate record commands. No configuration per session. Every turn, tool call, test run, and reasoning block is streamed to the team’s Capacitor server in real time, indexed for full-text search, and linked to the repository and pull request it belongs to. One simple setup, and your agents have memory.
Six core capabilities are available:
- Collaboration. Developers share live session links via tools like Slack. A teammate opens the link, reviews the spec or reasoning in progress, and contributes directly — without reconstructing context from message fragments.
- Multi-agent handoffs. Capacitor enables real-time coordination across multiple agents working in parallel or in sequence — handing off, handing back, and continuing work fluidly as the task evolves. Any agent can pick up exactly where another left off, with full context of what was done, what was decided, and what was already tried. Switch models mid-task, run agents in parallel on different workstreams, or bring in a second agent when the first hits a wall — Capacitor keeps the entire team, human and agent, in sync.
- PR review with session context. Review agents can pull the sessions behind a pull request at review time, giving reviewers and their agents access to the tests, attempts, and reasoning that produced the final diff. Changes become immediately explainable.
- Evaluations and institutional learning. Capacitor scores sessions against team-defined rubrics — correctness, test fidelity, surface area, time on task — and promotes durable findings into per-repository guidelines. Those guidelines are automatically loaded at the start of the next agent session, so lessons from one session become guardrails and best practices for every session that follows.
- Multi-player sessions. Capacitor turns solo agent sessions into team sessions. Launch any agent from the dashboard, share the link, and your whole team is in — driving, contributing, and building together in real time from any browser.
- Active session memory. Humans and agents can search the team’s full session history through built-in MCP tools. When a developer asks “have we worked on this before?” the agent searches, summarizes, and drills into the exact point where the relevant decision was made — surfacing prior context in the current chat instead of reopening closed questions.
What Teams Gain
Capacitor gives developers a new kind of leverage — one where every agent session makes the next one faster, smarter, and more informed:
- Products ship faster. The coordination overhead around agent-produced work doesn't shrink with Capacitor — it disappears. Reviews are grounded. Handoffs are instant. Reasoning is never lost. Developers stop explaining what happened and start shipping what's next.
- Agents get smarter. Before, every session started from zero. The work done in session 1 had no bearing on session 100. With Capacitor’s evaluation and guidelines system, findings from past sessions are automatically promoted into future ones. Teams that run Capacitor, even for a matter of hours, will have meaningfully better-performing agents than teams that don’t — not because the underlying model changed, but because the institutional knowledge accumulated.
- Reviews become reliable, not roulette. PR reviews that lack session context force reviewers — human and AI — to guess at intent. Confidently wrong review feedback is a real cost: it blocks correct changes, approves incorrect ones, and erodes trust in the review process. Capacitor gives reviewers the actual reasoning, not an inference from the diff.
- Agent choice. Because Capacitor normalizes sessions from major coding agents into a single event stream, teams can run the best tool for each job and switch when a better option emerges — without losing continuity. The history belongs to the team, not the tool.
- Agent work as a team asset. Today, a developer’s agent sessions are invisible to their teammates unless they actively share transcripts. With Capacitor, every session is immediately available to the team through the dashboard, searchable by any agent, and linkable from Slack or a pull request. Work done by one developer stops being siloed knowledge and starts being shared infrastructure.
Built on KurrentDB: Shared Session Memory
Capacitor is built on KurrentDB, Kurrent’s event-native database. Every agent turn, tool call, and session lifecycle event is stored as an immutable event in an append-only stream — the same architectural pattern that financial institutions and other regulated industries rely on for auditability and accuracy. This means Capacitor’s session history is a deterministic record that can be shared, replayed, queried, and projected into any view a team needs.
This foundation also makes Capacitor vendor-neutral by design. Sessions from Claude Code, Codex, Cursor, and other coding agents are serialized to the same event stream using the same schema. Teams are not locked into a single AI vendor, and switching agents does not mean losing history.
Industry Context
The launch of Capacitor coincides with a broader shift in how software engineering teams are organized around AI. According to the 2025 Stack Overflow Developer Survey, 84% of developers are now using or planning to use AI tools in their development process — with more than half of professional developers relying on them daily. As these tools become standard infrastructure rather than individual productivity shortcuts, the question of how teams coordinate around agent-produced work — how they review it, hand it off, build on it, and learn from it — has become a first-order engineering management problem.
Existing tooling does not address this. Session transcripts, where they can be saved at all, are static exports with no search, no team visibility, and no pathway for the lessons they contain to influence future agent behavior. Capacitor is the first purpose-built platform to treat coding agent sessions as durable, collaborative, and actionable team infrastructure.
“Our team was running Claude Code and Codex daily and losing critical context every time a session closed — reasoning gone, handoffs broken, decisions re-litigated from scratch,” said Alexey Zimarev, Kurrent’s VP of Engineering. “We started capturing sessions in KurrentDB, and it changed everything. We instantly knew that every team needs this, so we created Capacitor."
Availability and Pricing
Kurrent Capacitor is available today in private preview — free to use, simple to deploy, and compatible with Claude Code, Codex, and Cursor coding agents. Teams can request access immediately at https://capacitor.kurrent.io/. The Capacitor CLI is open source and available at https://github.com/kurrent-io/kcap-cli.
About Kurrent
Kurrent builds the AI-native memory infrastructure that powers the age of agents. Its event-native platform captures, preserves, and streams real-time, business-critical data with full historical context — giving both humans and AI agents the persistent, trusted foundation they need to act with confidence. From enterprise data pipelines to agentic coding workflows, Kurrent's technology is deployed in mission-critical use cases across finance, tech, oil and gas, manufacturing, retail, healthcare, automotive, government and many other industries globally. Kurrent is available on AWS, Microsoft Azure and Google Cloud Platform and as an on-premises solution. Headquartered in San Francisco. For more information, visit www.kurrent.io.
Contacts
Drew O'Brien
drew.obrien@kurrent.io
