-
-
Dashboard
-
PH #1 Product of the day
-
Task Worklog
-
Daily Summary
-
AI provider integration
-
Reddit viral post
-
Open source, privacy and local first principles
Inspiration
Engineering tools record planned tickets and final commits, but much of the actual work remains invisible: debugging, investigation, code review, interruptions, experiments, and context switching. AI coding agents have accelerated implementation while increasing the review and debugging burden. After interviewing more than 50 senior engineers and engineering leaders, we built Meridian to capture this missing layer of development work automatically for developers and engineering teams.
What it does
Meridian is a local-first, open-source AI work journal for developers and engineering teams. It runs quietly in the background on macOS and Windows, reconstructing work from activity across coding agents, IDEs, browsers, terminals, and related development tools.
It organizes activity into a meaningful timeline, connects work with relevant Jira and GitHub tickets, and generates:
- Daily work summaries
- Worklogs and stand-up notes
- Draft project updates
- A searchable record of completed and unplanned work
Users review and approve generated content before sharing it. Meridian requires no cloud account, keeps data on the device until the user chooses to share it, and is MIT-licensed for individual developers.
How we built it
We built Meridian with TypeScript, React, Rust, Swift, Python, Tauri, and SQLite. GitHub and Jira integrations use their APIs and OAuth to connect captured activity with project context. Bring your own AI provider like claude code, codex, cursor or Cloud/Local LLMs that help transform raw activity into structured summaries and updates.
We use PostHog and Sentry for product analytics and reliability monitoring, with GitHub Actions, Docker, and Kubernetes supporting development and delivery.
Challenges we ran into
The hardest challenge was converting fragmented activity into an accurate work timeline without overwhelming users with noise. We also had to make background capture reliable while preserving privacy and keeping processing local.
Other major challenges included AI provider connection failures, OAuth token-refresh race conditions, database corruption detection and recovery, telemetry stalls, daemon startup reliability, and maintaining stable releases across macOS and Windows.
Accomplishments that we're proud of
We launched Meridian on Product Hunt on August 17, ranking #1 Product of the Day and #3 Product of the Week. Product Hunt featured us three times across its daily and weekly newsletters, while our Reddit post about the problem reached 200,000 views and generated more than 1,000 interactions and strong positive feedback.
We built Meridian alongside 20 beta users, several of whom became heavy users and said they could no longer imagine writing daily updates without it. Within 10 days of launch, Meridian recorded more than 600 downloads and currently has 420 daily active users.
We are also proud that Meridian is MIT-licensed, local-first, and useful without requiring developers to move sensitive work data into another cloud platform.
What we learned
The most important lesson was that developer productivity cannot be understood from tickets and commits alone. The invisible work between those artifacts often explains delays, complexity, and real engineering impact.
We also learned that automation must remain reviewable. Developers want help reconstructing their day, but they need control over what gets shared. Privacy, transparency, and user approval are essential for earning that trust.
What's next for Meridian
Next, we will expand Meridian’s memory capabilities and develop an enterprise offering in collaboration with four engineering teams already signed on as design partners.
Log in or sign up for Devpost to join the conversation.