PDK Dev Call Updates – January 5th, 2026

[arminsabouri]

PDK Dev Call Updates – January 5th, 2026

Process / Meta

• The call cadence worked reasonably well, but with contributors across disparate time zones, we should keep this call lightweight and rely more on focused project-specific calls.

• Need a reliable note-taking AI for 1:1 calls. Gemini works adequately for capturing technical discussions.

• Agreed to bring back weekly sync calls with a stricter format.

  • New schedule:

    • Tuesday 10:00 AM EST -> weekly sync
    • Wednesday 11:00 AM EST -> mob programming / review (optional for everyone)

• Async updates are canceled.

• CI discussion:

  • We currently don’t run CI checks on every commit in a PR; there’s no strong historical justification.
  • Running lint and fmt on full commit history could help, but it’s not the highest priority.
  • Testing focus should shift toward fuzzing, adversarial testing, and malformed input handling (something chavic is already working on).

• Ben proposed making fuzzing the topic of an upcoming mob programming session.

Armin

• Working out the bip352 math for bip77 and multi party payjoins.
• Pairing with nothingmuch of some writing for the tx-indexer project. The hope is to write the project architecture which should help attract other collaborators.
• Review Async persistence #1235

Ben

• Helped Shehu get Nix-based CI checks validated and passing.
• Planning to rebase and land on other optimization work alongside Shehu’s changes (based on some suggestions from nothingmuch).
• Liana bumped MSRV to 1.85. Coordinating with their dev team on how to land database migrations required for the integration.

Chavic

• Working on primitives FFI error validation
• Receiver-side FFI error exposure is now unblocked and is the main focus for this week.

Spacebear

• Investigated directory health-check failures overnight; root cause was an expired certificate.

• Restored health checks after fixing certs; total downtime was ~6 hours.

• Pointed the main Payjoin domain to the ACME-enabled host. The old host is now updated and no longer uses Redis.

• Next step: review and land the unified service Unified payjoin service #1232 .

• File-based storage currently lacks metrics; adding metrics is the next coding priority for the unified service.

• Async persistence implementation is ready for review but still needs FFI exposure.

  • This may be blocked by pending UniFFI-Dart work and requires Spacebear review.

Nothingmuch

• Needs to catch up on code reviews.
• Focused primarily on writing and design work.

Zealsham

• Spent the past week peer-reviewing the async persister design with LDK developers.
• Providing integration guidance for LDK Node.
• Draft PR for LDK Node integration expected in approximately two weeks ™.
• Feedback from LDK: session persistence terminology is confusing.
• A dedicated blog post explaining persistence semantics would significantly help integrators.