In circulation
660
33 people Γ 20 β a closed economy
Open bounties
75
PRs awaiting a first review
Top reviewer
Sean Daley
359 πͺ of bounties across 212 reviews
Charity reviews
294
of broke authors' PRs Β· 5 people at zero
Top performers
What actually keeps the economy moving: the bounty value you reviewed, and the bounty value you shipped (PRs landed and plans published, whether or not anyone reviewed them). A big PR counts up to 3Γ, nothing is split, and charity reviews count just the same β so this won't match wallet balances, and that's the point.
π Top reviewers
total bounty value reviewed since the reset (before splitting)
1
SE
Sean Daley
πͺ 76.27
359 πͺ Β· 212 reviews
2
WI
Wills Bithrey
πͺ 47.47
241 πͺ Β· 135 reviews
3
MA
Martin Hartt
πͺ 22.5
209 πͺ Β· 100 reviews
4
DA
David Culemann
πͺ 3
206 πͺ Β· 121 reviews
5
AD
Adolfo Tamayo
πͺ 3
176 πͺ Β· 102 reviews
π’ Top shippers
total bounty value landed since the reset, reviewed or not
1
DA
David Culemann
πͺ 3
303 πͺ Β· 158 PRs + 1 plan
2
AD
Adolfo Tamayo
πͺ 3
172 πͺ Β· 86 PRs + 2 plans
3
GE
George Gebbett
πͺ 12.4
165 πͺ Β· 102 PRs + 2 plans
4
RH
Rhiannon Davies
πͺ 0
161 πͺ Β· 104 PRs
5
DA
Daniele Esposti
πͺ 20
155 πͺ Β· 102 PRs
Balances click a row for their coin history
Your wallet, not a score: a low balance usually just means you ship a lot. You earn by reviewing others' PRs and tech plans (approvals, change requests, and real comment reviews all count, split equally between reviewers); you pay when your own work gets reviewed. Broke? One review starts refilling.
| Who | Balance | Ξ | Earned reviewing | Paid for reviews |
|---|---|---|---|---|
| Aaron George (@zoonage) | 15.98 | -4.02 | 55.98 | 60 |
| Adolfo Tamayo (@adolfo-tamayo) | 3 | -17 | 93.22 | 110.22 |
| Alex (@alex-trosh-lawhive) | 18 | -2 | 0 | 2 |
| Andy Stevenson (@andys-lawhive) | 20.1 | +0.1 | 103.1 | 103 |
| Arnold#1a35e1 (@1a35e1) | 20.33 | +0.33 | 0.33 | 0 |
| bshelford | 17 | -3 | 0 | 3 |
| Cesar Faurat (@cesarfaurat) | 31 | +11 | 119 | 108 |
| Daniele Esposti (@danieleesposti-lawhive) | 20 | 0 | 64.85 | 64.85 |
| Dave Aitken (@actionshrimp) | 0 | -20 | 63.94 | 83.94 |
| David Culemann (@davidculemann) | 3 | -17 | 177.76 | 194.76 |
| Dominic West (@dom-lawhive) | 3.83 | -16.17 | 44.33 | 60.5 |
| Emma Bowen (@emma-lawhive) | 11.4 | -8.6 | 32.4 | 41 |
| Florian Zysset (@fzysset-lh) | 0 | -20 | 41.66 | 61.66 |
| Gareth (@GWMolyneux) | 27.17 | +7.17 | 7.17 | 0 |
| George Gebbett (@georgegebbett) | 12.4 | -7.6 | 48.31 | 55.91 |
| Harry (@Harrry-G) | 2 | -18 | 36.16 | 54.16 |
| Jaime van Oers (@jvanoers) | 65.97 | +45.97 | 63.97 | 18 |
| JosΓ© MarΓa Ruiz Aguilera (@josemariaruiz) | 0 | -20 | 0 | 20 |
| Kelan Littley (@kellit-lh) | 60.18 | +40.18 | 49.18 | 9 |
| Libaan Hassan (@Libaan-H) | 5 | -15 | 4 | 19 |
| Lukas Ondre (@lukasondre-lh) | 0 | -20 | 31.84 | 51.84 |
| Martin Hartt (@martinhartt) | 22.5 | +2.5 | 141.96 | 139.46 |
| Maxton Scott Coulson (@maxton-scott-coulson) | 5.89 | -14.11 | 67.89 | 82 |
| Moin Bukhari (@moinbukhari-lh) | 48.3 | +28.3 | 72.3 | 44 |
| Mostafa Yassin (@sh3raawii) | 2.5 | -17.5 | 42.5 | 60 |
| Peter Zachares (@zachares) | 27.2 | +7.2 | 65.98 | 58.78 |
| plproner | 19 | -1 | 0 | 1 |
| Pratik Borde (@sparkcanon) | 14.9 | -5.1 | 95.5 | 100.6 |
| Rhiannon Davies (@rhiannon-davies-lh) | 0 | -20 | 67 | 87 |
| Sean Daley (@Scdales) | 76.27 | +56.27 | 172.27 | 116 |
| Svjetlana (@svjetlanadjajic) | 20.66 | +0.66 | 0.66 | 0 |
| Wills Bithrey (@WillsB3) | 47.47 | +27.47 | 146.47 | 119 |
| Winston Angkajaya (@ston88) | 38.95 | +18.95 | 81.95 | 63 |
Open bounties
PRs and π tech plans awaiting their first human review, newest first, each showing its bounty (PRs: 1β3πͺ by changed lines; plans: 1β3πͺ by word count). The bounty splits between everyone who reviews, and you can π boost one with up to 3 of your own coins to get it looked at sooner, or π tip the author for the work itself. π€ means the author is broke, so reviewing is charity. Quiet for 14+ days = hidden, unless someone π boosts it back onto the board.
RH
π€
RH
π€
RH
π€
RH
π€
RH
π€
RH
π€
GE
πͺ
ZO
πͺ
CE
πͺ
RH
π€
RH
π€
HA
πͺ
data-dbt#4065
int_langfuse_v3_traces: incremental slim traces on a 14-day lookback
rhiannon-davies-lh
data-dbt#4064
test: warn when a lawyer's case count is summed across two platforms
rhiannon-davies-lh
data-dbt#4063
int_langfuse_traces: read the 78 GB input column once, not per consumer
rhiannon-davies-lh
data-dbt#4062
feat: settle the v3 matter reference as a string at the staging boundary
rhiannon-davies-lh
data-dbt#4061
Langfuse feedback reports: use the pre-extracted input_* columns
rhiannon-davies-lh
data-dbt#4059
int_posthog_v2_first_sessions: incremental on a 3-day lookback
rhiannon-davies-lh
platform-v3#14483
fix(security): fix confused deputy bug on lawyer matter note mutations
georgegebbett
infra#208
Add more GCP services and allow GCP SA name override
zoonage
data-dbt#4058
Terraform the firm-migration BigQuery datasets with tightened IAM (DM-59)
cesarfaurat
data-dbt#4057
dim_gb_clients: v3 identity wins over v2 and v1
rhiannon-davies-lh
data-dbt#4056
docs: platform v3 is all markets, not US-only
rhiannon-davies-lh
platform-v3#14482
feat(agent-framework): court-filing pre-flight in drafting, procedural research in legal-research β all variants (LEX-968)
Harrry-G
+ 63 more open PRs
Stacks in flight
Open PRs chained base-to-tip, the way Graphite builds them: a PR whose base branch is another open PR is riding on top of it. A stack is only visible while it's live, because once the bottom lands GitHub retargets the rest to main and the chain is gone. So this is what's in flight right now, not a season score. A green dot means someone has already reviewed that rung.
Deepest live stack
9 deep
platform-v3 Β· David Culemann
Stack adoption
64%
63 of 98 open PRs ride in a stack Β· 14 live stacks
π₯ Stackers right now
open PRs riding in a stack of two or more
1
DA
David Culemann
12 PRs Β· 2 stacks Β· 9 deep
2
GE
George Gebbett
8 PRs Β· 2 stacks Β· 6 deep
3
HA
Harry
8 PRs Β· 2 stacks Β· 5 deep
4
WI
Wills Bithrey
8 PRs Β· 2 stacks Β· 5 deep
5
DO
Dominic West
7 PRs Β· 1 stack Β· 7 deep
platform-v3davidculemann9 deep
#14455
fix(matter-service): only say a check passed when its outcome says so
1πͺ
#14456
fix(legal-os): stop the sidebar saying Passed when nothing said so
1πͺ
#14457
feat(matter-service): let AML and Source of funds fail
1πͺ
#14458
feat(matter-service): expect a client-level check for every client
2πͺ
#14460
feat(platform-api): tell matter-service which clients are still active
1πͺ
#14461
feat(legal-os): show client-level checks grouped by client
1πͺ
#14464
fix(legal-os): open a check's evidence from its chip
2πͺ
#14465
chore(matter-service): retire the dev seed script for matter checks
1πͺ
#14466
fix(legal-os): open task attachments through the shared hook
1πͺ
platform-v3dom-lawhive7 deep
#14278
feat(transaction-service): add Cost model and COST/EXPENSE enum values
1πͺ
#14292
feat(identity-service): add ContactLabel enum array to Contact (BOS-2585)
1πͺ
#14293
feat(transaction-service): derive cost totals with computeCostTotals
2πͺ
#14316
feat(billing): add billing.cost.get procedure and gateway (BOS-2588)
2πͺ
#14317
feat(billing): add billing.cost.list procedure and gateway (BOS-2587)
2πͺ
#14343
feat(billing): add billing.cost.create procedure and gateway (BOS-2589)
2πͺ
#14399
feat(billing): fill invoice lines from costs, publish exclusivity guard, void release (BOS-2592)
2πͺ
platform-v3georgegebbett6 deep
#14208
feat(matter-service): read many matters by id, access-filtered
2πͺ
#14209
feat(billing): a firm-wide pane listing every billing document
3πͺ
#14268
feat(billing): a billing row opens its matter from the matters table's pill
1πͺ
#14269
refactor(billing): a billing document's readers are the matters it bills
2πͺ
#14270
feat(billing): a firm-wide row opens its document in a read-only takeover
2πͺ
#14337
feat(billing): a draft previews from its record, not from a form
2πͺ
platform-v3martinhartt6 deep
#14429
feat(lawrence-api): fetch case-context citations and widen the CCO schema (LEX-910)
2πͺ
#14430
feat(agent-framework): teach agents to cite case-context citation events (LEX-910)
2πͺ
#14431
feat(agent-worker): emit usage, model and trace data parts from the TS chat stream (LEX-932)
2πͺ
#14433
feat(lawrence-api): remove the proactive next-steps message drafting path (LEX-934)
2πͺ
#14432
fix(dev-multi): merge the caller's PORTLESS_PRESERVE_VARS instead of overwriting it
1πͺ
#14446
feat(lawrence-api,lawrence-components): sidebar rows for load_skill and vfs_list (LEX-933)
1πͺ
platform-v3WillsB35 deep
#12785
docs(ui): plan UiStrings i18n seam for @lawhive/ui (UI-100)
1πͺ
#12787
feat(ui): add UiStrings i18n seam for @lawhive/ui (UI-100)
2πͺ
#12793
feat(ui): read UiStrings chrome via useUiStrings in the filter surface (UI-100)
2πͺ
#12795
feat(legal-os): localise @lawhive/ui filter chrome via UiStrings (UI-100)
2πͺ
#12797
test(ui): cover the UiStrings seam override behaviour (UI-100)
1πͺ
platform-v3Harrry-G5 deep
#14424
fix(lawrence-api,matter-service): tell the agent a document was finalised instead of "not found" (LEX-962)
2πͺ
#14425
fix(lawrence-api): search follows list cursors and reports failed or truncated namespaces (LEX-970)
2πͺ
#14480
fix(legal-sources): port the OpenLaws fallback ladder and regulatory ranking to the TS adapter (LEX-967)
2πͺ
#14481
fix(agent-worker): glob normalisation + case-insensitive matching; stat passes finalisation metadata end-to-end
1πͺ
#14482
feat(agent-framework): court-filing pre-flight in drafting, procedural research in legal-research β all variants (LEX-968)
1πͺ
platform-v3zachares5 deep
#14443
feat(agent-framework): the pure half of context compaction (LEX-931)
3πͺ
#14444
feat(agent-framework,agent-worker): the compaction decision and its summariser (LEX-931)
3πͺ
#14445
feat(agent-worker): run compaction on the chat turn, and rebuild from it next turn (LEX-931)
3πͺ
#14474
fix(agent-framework): compact against what the serving path sustains, not the model spec (LEX-965)
1πͺ
#14475
fix(agent-worker): a token-count failure must not silently skip compaction (LEX-965)
1πͺ
data-dbtrhiannon-davies-lh4 deep
#4059
int_posthog_v2_first_sessions: incremental on a 3-day lookback
1πͺ
#4061
Langfuse feedback reports: use the pre-extracted input_* columns
1πͺ
#4063
int_langfuse_traces: read the 78 GB input column once, not per consumer
1πͺ
#4065
int_langfuse_v3_traces: incremental slim traces on a 14-day lookback
1πͺ
Who's funding whom
Net flow = coins earned reviewing minus coins paid for reviews of your own PRs. Green earns more than they pay; red pays more than they earn.
Race over time drag the scrubber, click a chip to spotlight
Every move since the reset, replayed: balances, reviews given, or PRs shipped. Press play and watch the top eight race along the timeline, everyone else riding in grey.
Recent coins
The rules
- Everyone starts with 20 coins.
- Reviewing a PR or a tech plan earns from its author: approvals, change requests, and comment reviews with substance all count. Rubber-stamping is not the way to earn.
- The bounty is sized by the work: PRs pay 1β3 πͺ by changed lines (200 / 800), plans pay 1β3 πͺ by word count (2,500 / 5,000).
- It's split equally between everyone who reviews β no race, no rushing. A PR's bounty is paid when the PR lands, to everyone who reviewed it before then; a tech plan pays each share as that review lands.
- Short on coins? You pay what you can. Only an empty balance makes it charity.
- Need something reviewed sooner? π Boost any open bounty with up to 3 of your own coins; reviewers collect them on top.
- Broke? Reviews of your PRs are charity: welcome, but they earn the reviewer nothing.
- Bot reviews are ignored. They never earn and never charge, so automation can't drain the economy.
- Coins are never minted or burned. It's a real economy.