All Core Devs - Testing (ACDT) #83, June 15, 2026

Protocol Calls & happenings
,
1

Agenda

Shared EL-CL Topics

We will have breakouts today…

Glamsterdam

Prep for fast resolution at ACDE #239.

ePBS - EIP-8282 builder deposits (go / no-go?):

Other Topics

  • SSZ Engine API (EL-side) @barnabasbusa - status + what is blocking @MariusVanDerWijden’s execution-apis#793 (engine Rest-SSZ spec); get client teams implementing. Earlier ask: needs a 3rd CL volunteer.
Devnet Status / Readiness / Scheduling
  • bal-devnet-7 - the last EL-specific devnet; EL benchmarking.
    • New release today: tests-bal@v7.3.2, includes forks <= Prague; no change to Amsterdam coverage.

EL Breakout

Gas Repricings:

Other topics:

CL Breakout

ePBS - fork choice (EIP-7732), @jihoonsong - consensus-specs#5348 Modify get_proposer_head for Gloas

  • get_proposer_head takes/returns ForkChoiceNode; is_parent_strong now counts a PENDING parent’s support (#5305) - both Gloas. Also drops is_shuffling_stable (Fulu proposer lookahead; not Gloas-specific).
    • Review + interop implications for the ePBS devnet?

Other topics:

  • QUIC vs mplex (CL networking) @barnabasbusa - aim to decide in the coming weeks whether to drop mplex for beacon p2p. Teku shipped QUIC in its latest stable release; status fro Nimbus? (ACDC #180 had removal at ~Gloas + 2 months; Barnabas wants the decision sooner.)

  • Builder API / buildoor readiness (gates ePBS + 8282 testing): prysm Builder API + auth working @terencechain; (beacon-APIs#588) @nflaig.

  • Staked Builder API builder-specs#138 merged. Heads up.

Last call:

Please comment on this issue with other topics!

Meeting Time: Monday, June 15, 2026 at 14:00 UTC (60 minutes)

GitHub Issue

1 Like
2

Video, transcript & chatlog

News coverage

Resources

3

Meeting Summary:

This meeting was ACDT issue number 83 on June 15th, 2026, focusing on the proposal 8282 for builder execution requests and related execution layer (EL) changes. The main discussion centered around introducing a new system contract to handle builder deposits instead of verifying them at the slot level, which would reduce signature verification load and prevent commingling between validator and builder deposits. Participants including Justin, Nico, and Berlin explained that this approach would be cleaner and less complex to implement than alternative optimizations, though it would require additional testing and delay the Glamsterdam fork. The group debated whether to proceed with this change now or delay it by three months, with concerns raised about the complexity of system contract auditing and the potential impact on gas limits. The discussion also touched on the possibility of splitting EL and CL changes, with some suggesting shipping EL changes without waiting for CL if there was significant delay. The conversation ended with plans to continue these discussions in Thursday’s ACDE meeting and to proceed with Glamsterdam DevNet 6 planning, targeting next Wednesday for launch.

Click to expand detailed summary

Dan opened ACDT issue number 83 on June 15th, 2026, and outlined the agenda which included discussions on adding builder deposit system contracts and an execution requests engine API extension, as well as various repricing topics. The meeting began with introductions and technical checks, followed by Dan mentioning that there would be breakouts for the repricing topics. The discussion started with addressing the 80 to 82 proposal that was flagged in ACDC the previous week.

The team discussed EIP-8282, which proposes a new system contract to handle builder execution requests separately from validator deposits to improve scalability. Justin and Nico explained that this approach would avoid commingling between validator and builder deposits and reduce the need for signature verifications per slot. The group generally agreed this was the right solution to the deposit/scaling problem, though concerns were raised about the complexity of the system contract and testing requirements. The team aligned on moving forward with this proposal for Glamsterdam, with Barnabas suggesting to test it on glam-devnet-7, and agreed to estimate testing costs on the steel side before making a final decision at the upcoming ACDE meeting on Thursday.

The team discussed a performance solution for signature processing, with Enrico and Terence agreeing it represents the cleanest approach despite potential attack vectors involving high ETH costs. Berlin raised concerns about gas limit impacts, noting that while current limits of 193 million gas are manageable, higher limits could allow 16,000 builder deposits per slot instead of the tested 8,000. Potuz emphasized that the proposed solution is the right design choice regardless of specific attack concerns, arguing that the discussion should focus on whether to implement this design now rather than debating the particular attack scenario.

The team discussed implementing an EIP change, with consensus emerging around prioritizing correctness over speed. There were concerns raised about handling builder deposits before the fork, with some suggesting the mechanism could work similarly to withdrawals and consolidations. The group agreed to further discuss these implementation details in the upcoming ACDE meeting on Thursday, where they will also address potential delays to the Glamsterdam fork due to contract auditing requirements.

The team debated increasing the gas limit, with Potuz and Barnabas arguing against it without EPBS (Enhanced Propagation and Block Sizing) due to concerns about block propagation times. Berlin suggested increasing the gas limit to 110 million, but the discussion revealed ongoing disagreements about timing and propagation requirements. The team agreed to continue the discussion asynchronously before Thursday’s meeting, where they will consider proposals for two potential routes including the separation of EL and CL.

The team discussed plans for launching Glamsterdam DevNet 6, with initial plans to include EIP-8282 but concerns about the timeline impact. Barnabas suggested waiting unless ELs feel significantly delayed, while Dragan confirmed IL side changes are nearly complete with only minor padding items remaining. The group agreed to aim for launching DevNet 6 the following Wednesday, with the possibility of including EIP-8282 pending final review and security assessment. The conversation ended with plans to move to breakout sessions for further detailed discussions.

Next Steps:

  • dan: Drop the agenda and relevant links in the chat for the meeting.
  • Nico: Point out and discuss the scenario of handling builder deposits already coming in before the fork, and how EL would handle that.
  • dan: Make a better effort at estimating the testing cost for the system contract changes on the steel side.
  • All EL devs: Review the updated execution APIs PR (793) and adapt their implementations to the new spec.
  • Spencer: Aim to wrap up the test cases for the remaining EIPs by next Monday.
  • dan: Drop the tracker for Glamsterdam DevNet 6.
  • All participants: Comment on the ACDE issue with any reservations or thoughts before Thursday’s meeting.
  • All participants: Prepare thoughts and proposals for the two routes (splitting EL and CL, or not) for discussion on Thursday.
  • Barnabas: Create breakout rooms for the meeting.

Recording Access:

4

YouTube recording available: https://youtu.be/3OjZPPkrFq0