Evidence-0001: Promotion to Rank 1

Member details

• Matrix username: @guillaume:parity.io
• Polkadot address: 13psJuWEjBuZGaFqXvFnLMC6ME8RVVfQAhtFhydYjW45oKgZ
• Current rank: 0
• Date of initial induction: 2025/01/09
• Date of last report: N/A
• Area(s) of Expertise/Interest:
  • polkadot business-logic (aka the 'runtime')
  • the internals of the frame pallet framework
  • runtime and host APIs

Reporting period

• Start date: 2024/08/01
• End date: 2025/01/09

Evidence

Past work before the reporting period: I contributed to the internals of the frame pallet framework by writing the FRAMEv2 pallet macro (updating from decl_module! to #[pallet] macro: https://github.com/paritytech/substrate/pull/6877). I designed and implemented the instantiable pallet concepts (https://github.com/paritytech/substrate/pull/1800), this allows multiple instances of the same pallet in the runtime, bringing greater flexibility to pallets. I implemented the counted storage map (https://github.com/paritytech/substrate/pull/9125), and the iterable storage map without linked list (https://github.com/paritytech/substrate/pull/4185). Those features ease development of new features in polkadot pallets.

Work for the reporting period: I fixed tasks for instantiable pallets (https://github.com/paritytech/polkadot-sdk/pull/5194), further enhancing pallets flexibility.

In the substrate runtime primitives I contributed to the transaction extension implementation, this feature allows more flexibility in the runtime and enables us to deliver more features in polkadot system chains:

• I fixed implication order for tuple of tuple in transaction extension (https://github.com/paritytech/polkadot-sdk/pull/7028) making it easier to use the feature.
• I reviewed the PR and identified:
  • the PR was missing the pipeline for bare extrinsics, so weight wasn't checked nor registered for bare extrinsics: https://github.com/paritytech/polkadot-sdk/pull/3685#discussion_r1794069271
  • CheckNonce prepare was overriding the whole storage item with value coming from validate, potentially erasing other change in the account storage: https://github.com/paritytech/polkadot-sdk/pull/3685#discussion_r1698220569
  • The pipeline was made to allow multiple version of transaction extension but there was no plan to make the version part of any signed payload: https://github.com/paritytech/polkadot-sdk/pull/3685#issuecomment-2400404179

I also contributed to some cumulus runtime components, on storage weight reclaim:

• I found and fixed a minor bug in the reclaim calculation (https://github.com/paritytech/polkadot-sdk/pull/6140)
• I updated the transaction extension to wrap all transaction extensions, to provide a more accurate reclaim (https://github.com/paritytech/polkadot-sdk/pull/5273), this may improve parachain throughput.

In the polkadot runtime I found out the benchmark configuration was suboptimal, potentially overestimating some weights, and fixed it (https://github.com/polkadot-fellows/runtimes/pull/525).

Voting record

Misc

• Question(s):

• Concern(s):

• Comment(s):