Code:
Name: LRP - Luke
Term: 3 Cycles
Cycle Amnt: 20,000
Total Amnt: 60,000
Author: JSKitty
Receiver: Luke
Address: DNkAdmxHiGvYvMV3Cvtb4FFYhJMUTc8oR2
Created: 15-10-2024
Status: Active
Vote Hash: 5081ffdf9adc52f886aaec6b93286228a7ed360035055d2bebadd7c5b4014f57
Role and Introduction
Luke is the primary backend developer across the majority of central Labs Platforms under JSKitty, he develops PIVCards, Labs Oracle, our Node Maintenance systems, and assists with architectural work and code-reviews for other parts of Labs - Overall, he's a core member of Labs' infrastructure, keeping everything running smoothly while adding new features and expanding the ecosystem.
The Park Fire
Luke unfortunately had more than PIVCards upgrades to deal with this cycle - with the Park Fire to deal with, causing a temporary evacuation (a few weeks) with minor property damage.
As expected, some additional work that could've been completed in the same timeframe may have been delayed due to the circumstances at hand.
Progress Report
PIVCards: database upgrades, order scalability upgrades, cashback preparation, and more!
Luke's work on PIVCards has primarily been around backend-level database and scaling improvements - with PIVCards handling more volume than ever, his improvements will come in handy in the very near future, with all upgrades being finished, simply awaiting deployment in the staging deployment line (said upgrades require a LOT of testing to ensure the migration goes smooth, and the rewritten order scaling system is functioning perfectly, given the sensitivity of it).
A list of high-level improvements being:
- PIVCards Cashback: backend functionality is now complete, ready for frontend implementation, and subsequent feature launch.
- "Double-Trouble" patch: a fix for a quiet long-standing issue in which PIVCards could pay the upstream supplier twice, bleeding funds (fortunately, these were recovered).
- Database Upgrade: the PIVCards database has been upgraded to the new Labs standard, much more efficient, and highly data-redundant.
- Additional card delivery fixes: various extra card brands that had issues with delivery, are now further smoothed out.
All of the planned improvements from the prior proposal's plans have been implemented, perfectly on-target.
Labs Oracle: database upgrade and stability improvements!
The Labs Oracle database upgrade has been complete and deployed live on Production - you'll notice the charts on the Labs Oracle website (and API) now load substantially faster, thanks to Luke's implementation of an efficient chart database format.
PIVX Rewards: database upgrade!
Alongside PIVCards and Labs Oracle - the PIVX Rewards platform has it's database upgrade in the queue, this will be coordinated alongside the Labs Ads deployment, given that Labs Ads introduces new elements to the database schema that Luke has to wait for.
Proposal Plans
These plans are now ordered by priority, with the top-most items receiving the most attention and lower-most being "spare-time" ventures.
Plans may also change or expand during the proposal's timeframe, all work in which will be documented by PIVX Labs and reported on accordingly.
PIVCards: Internal Search Engine + Catalogue system
The PIVCards backend itself is now incredibly stable - however, our reliance on our card provider for both the Search engine and the Card delivery system means that, in recent times, the majority of downtimes have been outside of our hands, with no choice but to wait on their recovery.
With this, the plan to build an internal Search Engine for PIVCards is underway - we'll also be introducing a catalogue system with it, allowing users to explore and sort cards by Category.
Need to buy food? Scroll the food section.
Need a fuel topup? Gas refills at a dozen stations are a scroll away.
Hotel? PIVCardo!
Labs Oracle: Additional data sources
New exchanges and price aggregators are being added to the Labs Oracle backend for even more data redundancy - while these won't be exposed on the frontend or API, the addition of extra sources means that Labs Oracle is more accurate, real-time, and resilient to even the worst of 3rd-party outages.
Fun stuff: additional, low-priority free-time tasks
In spare time, Luke has been preparing hardware, models and code for an in-house AI model system that could potentially allow Labs to offer AI model inference services paid in PIV.
While it's not a primary objective this cycle, there is potential for us to progress on building a system allowing anyone to purchase AI credits in PIV, for private, instant access to uncensored, awesome LLM models.
If progress is made on such a project, and especially if the community is interested - we'll update more on this in the near future.
Last edited: