What did you build?
A from-scratch redesign and rebuild of the Livepeer Explorer, and the subgraph work underneath it. The main pieces:
-
Historical earnings data in the subgraph (livepeer/subgraph#217): the subgraph couldn’t say what a delegator’s stake is now or was at any past round, or what they earned in a given round, without calling the contract. I added cumulative reward and fee factors matching the contract’s own maths, delegator shares and snapshots, and orchestrator commission tracking, so all of it can be read straight from the subgraph. This is what makes the portfolio, earnings history and export possible.
-
Portfolio: track any mix of wallets, Safes and ENS names, with stake, rewards, fees and realised APR per round, plus a CSV export valued in USD for taxes.
-
Alerts: the Explorer flags anything you should act on, from an orchestrator missing reward calls or raising its cut to open votes, pending Safe signatures and stake ready to withdraw.
-
Transfer stake: move delegated stake between wallets without unbonding. The protocol has long supported this, but the Explorer never offered it.
-
Orchestrators ranked by what delegators actually earned, with full profiles: reward calls, cut history, delegators and commission.
-
Smooth transactions: balances refresh on their own once a transaction is indexed, a stake transfer the contract would reject is caught before you sign, with the reason shown, and Safe owners approve and delegate in one signature.
-
Built on the v2 Livepeer UI design system by Peace Node, the same one the upcoming Agent Console and livepeer.org are built on, so all three share one design language.
Built with Claude Code. I set the direction and design and reviewed every change. It has 165 automated tests, CI, CodeQL and dependency scanning, and has been through separate initial security reviews by Claude and Copilot, with all findings resolved.
Why does it matter?
The Explorer is where Livepeer’s 2,000+ delegators manage nearly 30M LPT (over half of all LPT), and where 100 orchestrators are judged when delegators choose who to stake with. It shows network data well, but it can’t answer a delegator’s real question: how am I doing, and should I do anything?
- Delegators can’t see what they earned. Per-round earnings history has been requested since 2022 (#143) but never landed, because the subgraph couldn’t provide a delegator’s stake or earnings over time. Holders with several wallets have no combined view, and nothing warns them when their orchestrator stops calling reward or raises its cut.
- This is the Delegate track’s own goal. The track aims for delegators who understand their rewards and know when moving stake would earn more, measured by funnel conversion. The new Explorer is built around exactly that, and its analytics match the current one’s, so the two can be compared on real traffic.
- It’s a simpler foundation. Since 2019, the codebase has grown by adding new tools on top of old ones, so it now relies on several overlapping libraries and an older design system. The new Explorer does more in about half the code and shares the v2 design system with the upcoming Agent Console and livepeer.org.
Link to the work
- Explorer PR: app: rebuild the explorer from scratch as a portfolio-first app by adamsoffer · Pull Request #828 · livepeer/explorer · GitHub
- Subgraph PR: feat: cumulative factors and delegator shares for O(1) stake computation by adamsoffer · Pull Request #217 · livepeer/subgraph · GitHub
- Subgraph deployment (with these changes): Livepeer Staging | Graph Explorer
- Preview: https://explorer-arbitrum-one-git-explorer-v2-livepeer-foundation.vercel.app?_vercel_share=D0monoZAf7AngNAhlLNhGZnvNxkWZP70
- Rollout: launching next at https://beta.explorer.livepeer.org, with the aim of replacing the current Explorer at explorer.livepeer.org once it has passed an additional internal security review.
Evidence of impact
None yet in usage terms, since the new Explorer hasn’t launched publicly. Its analytics match the current Explorer’s (#722), so the two can be compared on delegation funnel conversion, the Delegate track’s own measure of success: at first on beta traffic, and fully once it replaces the current Explorer.
Community proof points
- Rick Staa (Technical Director): “I like the direction: the expanded feature set, the more modern design, and the increased consistency across the different platforms. I don’t have any major objections to moving toward this new design system and tooling.”
- Elliott Conway (Delegate track owner): the portfolio approach “matches our analysis results for Delegator UX, which was planned to be implemented in the current SPE.” This application is submitted at his request.
- Rich O’Grady (Ecosystem Director): “What Adam has built is very good.”
Additional feedback from orchestrators and delegators who try the preview is welcome below.
Amount requested
$6,500 USD-equivalent in LPT:
