diff --git a/committee-steering/meetings/021-2024-10-31.md b/committee-steering/meetings/021-2024-10-31.md new file mode 100644 index 00000000..6bebc808 --- /dev/null +++ b/committee-steering/meetings/021-2024-10-31.md @@ -0,0 +1,424 @@ +# Akash Network - Steering Committee - Meeting #21 + +## Agenda +- Address community feedback and grievances. +- Review updates to project boards. +- Share updates on recent and upcoming events. +- Discuss progress on developer onboarding initiatives and YouTube content. +- Provide updates on the Cosmos SDK 47 upgrade and related technical developments. +- Review and discuss active governance proposals. +- Open floor for additional topics or feedback. + +## Meeting Details + +- Date: Thursday, October 31st, 2024 +- Time: 11:00 AM PT (Pacific Time) +- [Recording](https://h5r73bah365pqfr4x62x5sdut5px4wpprsnd6lu2hcej3x42jkxa.arweave.net/P2P9hAffuvgWPL-1fsh0n19-We-Mmj8umjiInd-aSq4) +- [Transcript](#transcript) + +## Committee + +- Anil Murty, Overclock Labs +- Artur Troian, Overclock Labs +- Greg Osuri, Overclock Labs +- Scott Caruthers, Overclock Labs + +## Participants +- Tyler Wright +- Adam Wozney +- Amanda Keay +- Andrew Gnatyuk +- Anil Murty +- Artur Troian +- Cheng Wang +- Damir Simpovic +- Ed Prado +- Fico Calderon +- Garrett Minks +- Joao Luna +- Juan Pablo Arce +- Robert Del Rey +- Rodri R +- Scott Carruthers +- Scott Hewitson +- Zach Horn +- Zeke Ezagui + +## Meeting Notes + +### Community Feedback +- Tyler Wright opened the floor for feedback and grievances from the community. +- Participants were reminded to share their input during meetings or reach out to core team members or insiders between sessions. +- Focus on making SIG operations and community contributions more transparent and accessible. + +### Project Boards Review +- Tyler emphasized the importance of the public-facing project boards. +- Boards track initiatives across product engineering and marketing, offering a transparent view of Akash’s ongoing efforts. +- Team members traveling for events have been contributing to updates despite logistical challenges. + +### Event Updates +- Amanda Keay provided updates on recent and upcoming events: + - **Permissionless Conference** in Salt Lake City: + - Akash team had speaking engagements and hosted a successful booth. + - Generated high-quality leads, particularly from AI developers, comparable to engagement levels at Token2049. + - **Cosmoverse in Dubai**: + - Akash led the AI track, hosted an AI-focused afterparty with over 300 attendees, and secured prominent branding. + - Team members participated in panels, discussions, and networking activities. + - **NeurIPS 2024** in Vancouver: + - Plans include a booth, speaking engagements, and workshops at a nearby venue, focusing on Akash's AI positioning. + - Attendees will have access to Akash Compute credits and networking opportunities. + - **ETH Denver 2025** preparations are underway, with plans to continue building Akash’s presence in the Web3 space. + +### Developer Onboarding and YouTube Content +- **Zealy Campaign**: + - Sprint 3 of the second campaign completed, with rewards distribution scheduled. + - A new onboarding process is being developed, focusing on guiding developers through a structured user journey from basic engagement to advanced contributions. + - Best campaign submissions, including guides and designs, will be compiled and shared. +- **YouTube Content**: + - Steering Committee meeting recordings are being prepared for upload. + - Upcoming focus will be on creating technical tutorials for deployments, Akash Console features, and GitHub contributions. + +### Cosmos SDK 47 Upgrade and Technical Updates +- Artur Troian provided updates on technical progress: + - The initial testnet for Cosmos SDK 47 is live, focusing on stability before expanding participation. + - Ongoing provider-side enhancements for Akash Console 2.0 include contributions from the community. + - Improvements to inventory exporter functionality were completed with community support. + +### Active Governance Proposals +- **[Rair Protocol Proposal](https://github.com/orgs/akash-network/discussions/709)**: + - Garrett Minks and Ed Prado presented the integration of Akash as a decentralized compute layer for deploying DApps. + - The goal is to move away from centralized solutions like GCP and offer Web3-native compute options. + - Proposal outlines a phased approach, starting with configuration migration and progressing to full feature parity with GCP-based setups. + - Community members were encouraged to review the proposal and provide feedback on GitHub. +### **[Provider Incentives Pilot 2 (PIP-02)](https://github.com/orgs/akash-network/discussions/696)** +- Presented by Cheng Wang: + - PIP-02 doubles the funding size from the previous pilot to expand GPU capacity and address rising demand. + - GPU models, including A100s, RTX 4090s, and future-ready 5090s, were detailed in allocation plans. + - Zeke Ezagui raised questions about: + - Flexibility in upgrading to 5090 GPUs when available. Anil clarified that allocations would be reviewed case-by-case to accommodate next-generation hardware while balancing commitments to current GPUs. + - The inclusion process for community providers in the incentive program. Cheng and Anil confirmed a manual vetting process based on outlined criteria, emphasizing fairness and adherence to quality and availability standards. + - The need for clear instructions for community providers to apply. Tyler committed to sharing detailed steps and ensuring community awareness. + - Anil reiterated that existing customer workloads would guide allocation decisions, with options to transition customers to upgraded hardware where feasible. + +### [Governance Proposal Best Practices](https://github.com/orgs/akash-network/discussions/170) +- Tyler Wright highlighted updates to governance proposal guidelines: + - New guidelines aim to streamline the process for submitting and reviewing proposals. + - Community members were encouraged to refer to these guidelines for leveraging the community pool effectively. + +### Open Floor Discussions +- Tyler encouraged participants to raise any additional topics or questions. +- Updates on SIG meetings and regular working groups were shared, including: + - Zealy meetings, focusing on developer onboarding. + - Akash YouTube working group, prioritizing technical content development. + - SIG chain meetings to discuss the Cosmos SDK 47 upgrade. + - SIG provider discussions on ongoing efforts related to Akash Console 2.0. + +## Action Items +- Tyler Wright will continue gathering feedback and ensuring transparency in SIG operations. +- Amanda Keay will finalize event plans for NeurIPS 2024 and ETH Denver 2025. +- Robert Del Rey will complete Zealy rewards distribution and redesign the onboarding process for developers. +- Artur Troian will stabilize and expand testnet participation for the Cosmos SDK 47 upgrade. +- Community members are encouraged to review and provide feedback on the Rair Protocol proposal and vote on PIP-02. +- Governance proposal updates and best practices will be shared to guide the community on effective use of the community pool. +- Announcements regarding SIG working group schedules and opportunities for engagement will be shared via Discord and GitHub. + +# **Transcript** + +Tyler Wright: All right, welcome everybody to the Akash steering committee. It is October 31st, 2024. during these monthly meetings, the steering committee gets together talks about everything that's happened over the course of that month in terms of SIGs, working groups and beyond. we follow a typical agenda which I will move over to now. But again, if anybody has any specific items that they want to discuss as a part of steering committee, again, please drop them in. We usually try to have these things in advance because oftent times we have a full plate of agenda items. But again, if there's anything specific that anybody wants feel free to drop it in here now. All right. Again, you can look at notes from the past meetings here. + +Tyler Wright: notes, transcripts, recordings for all the steering committees since January 2023 are available here. Again, one of the things we start off by doing is reviewing, discussing, and addressing any community feedback, grievances, or conflicts. This could be anything you see in SIG's working group structure. Anything that you think can be improved on, feel free to raise your hand and we can discuss those things right now. Cool. again, if anybody sees anything between meetings, feel free to reach out to myself or any member of the core team, insiders, etc., and we can again continue to u improve the process, make it easier for folks to get involved, etc. All right, continuing one of the big items that we discussed during these, steering committee calls is, the project boards. + +Tyler Wright: again, these are transparent public facing boards that anybody from the community can kind of track what's going on across community marketing initiatives as well as product and engineering. you can see our road map and these items are discussed at steering committee as well as other SIGs and working groups and continually updated by members of the core team and other members of the community that are working on specific efforts. So if you're ever interested in what's on the future of the road map across again product and engineering, again, please look out. This is always constantly being worked on and constantly being updated. I know some members of our community and marketing team have been working hard and have been on the road over the last month. and it's just been again a tireless amount of work that's been done. + +Amanda Keay: + +Tyler Wright: I do want to kick it off by handing it over to Amanda to just talk a little bit about some of the events that have gone on over the last month, including Cosmoverse and others, as well as a look forward to again some of the events that are on the horizon. + +Amanda Keay: Yeah, thanks Tyler. + +Amanda Keay: So about a month ago, I'll start three of us from the core team were able to go to permissionless in Salt Lake City. We had a couple different speaking engagements and an official side event during that week. it went really well. I felt like the conference had a lot of really high quality people that came by, a lot of specifically AI developers that were interested in talking with us. + +Amanda Keay: Zach and I were manning the booth and I think as far as quality leads and the amount of leads, it was up there with the amount total that we got altogether going to token. So, it was a really good success for us there. and then we just a week or so ago got back from Cosmoverse in Dubai. That event was really good. a three-day event. Greg, we actually led the AI track at Cosmoverse. So, we were able to have our branding front and center as well as multiple speaking opportunities by both Greg, Anneil, and Zach. So, that was really awesome. We also hosted a AI afterparty following our AI track day just right next door. we had 300 plus attendees show up. + +Amanda Keay: So we were able to really flex our branding there and just prove that Akos branding is as strong as it is right now. So that was really great. Looking forward to the next month. We don't have any travel per se in November. A few hackathons that we might be visiting this month but then we're really focusing our efforts on Nurups which is in Vancouver this year from the dates November through the 14th. So Akos will be doing a larger booth at the main event as well as Greg got accepted to be a speaker on the main stage and we are planning on doing a social house just right down the road a really cool venue used to be an old bank so we have that space for two days we're going to have some workshops some special speakers we're finalizing that now as well as just a space for people to come and network and talk and if they wanted to hack they can have + + +### 00:05:00 + +Amanda Keay: Akos compute that we're going to give credits for and just kind of build our branding in the more AI space. looking forward to that doing a lot of planning. so that is taking a lot of time right now and I think that will put us through to the end of the year and then already honestly working on ETH Denver events which is in February. So things are moving rather quickly and it's exciting. + +Tyler Wright: Thank you, everybody that's been involved in all those events over the last month. And as we look forward, as a reminder again, there's been a lot of discussion in GitHub around Q4 proposal and now there's a vote on chain. So, if you have the ability to vote, please do so. as we continue to again, fine-tune and, continue to build the brand that's already very strong. So, much appreciate everyone's time. Does anyone have any questions for Amanda about anything events related? I know she has to head out very soon to some other obligations. Cool. All right. + +Tyler Wright: we talked about it in again a weekly working group around the website and docs. again there's folks that contribute to console. If you are contributing in any way via the states and working group structure please make your contributions known via comments inside the specific area in which you're again contributing. this helps with transparency and making sure it's easy for folks that are contributing to get rewarded for those contributions. + +Tyler Wright: so again just want to make that u clear to everybody that you can look on this board. again get involved in the signing working groups. We discuss specific issues that need solving inside each of the repos and then again if anybody can get involved in the community we look for those contributions. So thank you to all that are getting involved. I did see that Robert had some items specifically that he wanted to discuss on the community and marketing side. So, Robert, I'll give you the four to quickly talk about Zilly and the Kosh YouTube working group. Yes. + +Robert Del Rey: Thank you, Can you guys hear me? All Really cool. so yeah, as you some of you may know, we are leading Pyber and I are leading the Aash dev on boarding campaign on CV. It's a platform that helps us create missions and make it look like a game to engage with a cashache. this is the second three month campaign. So we basically do six months of campaign on silly engaging with more than 2,000 people which is awesome. the spring number three of the second season ended just yesterday. So right now we are just doing the math to determine the amount of rewards that are going to be paid out. We're going to start paying out those rewards on Monday. + +Robert Del Rey: you can also expect our city working group call happens every So on next Monday in our call you can also expect the project status In that report we add information analytics like we add how many people are engaging with the platform and more stuff that we share with the community. So you can see how we are doing and how is Aash benefiting from this. So really cool. it was wild trying to manage this while on the road during Cosmoverse. but we managed to do it. very happy for the support I got from Benji and also more people that engaged with this. We had a really good call on during Cosmover. I actually share the screen. + +Robert Del Rey: Sorry, I share my camera during verse and 20 people joined that city working group call and they were very happy to see our aash booth in Cosmo verse and got really excited. one thing that I can tell you is that yes, the campaigns are done in editing, but we created a document and we compile of the best submissions that we got on Cely from guides and videos to designs all those of the best really really good quality submissions we have them in a document and we're going to share those with you also very soon. + + +### 00:10:00 + +Robert Del Rey: we will give some extra rewards to those participants going the extra mile. So as for now on November the first two weeks of November only expect the rewards being paid out but no more sprints at the moment. we are actually redesigning the process and we want to make it look more like a user journey like the next phases of cely or whatever tool that we decide to use to do this. + +Robert Del Rey: we want to make it look like a journey where people get to know a cache, they join the social media, read the documentation, they start deploying and doing stuff on a cache and then at the end we show them how to get involved in GitHub like a star when you put a start in one of the repos to get all the updates there's going to be mission about that so people can start the cache GitHub + +Robert Del Rey: + +Robert Del Rey: stuff and yeah so yeah it's going to be really nice we're going to go back to the roots of silly on boarding new developers we are very thankful for the people participating every month in the sprints but with this user journey we want to go back to the initial goal of this campaign which was attract new developers to a cache so it's a concept we're still developing it so that's the work we're going to do in the following weeks hopefully and release that to the public in the next two weeks mid November. + +Tyler Wright: Perfect. Thank you. + +Robert Del Rey: So that on ceiling. Let me know if you guys have any question about it. One, two, three. No, that's okay. you can also drop any questions in the chat or go directly to the city chat for more stuff. in regards to Akash YouTube one or two months ago we started the Aash revamping efforts we believe the Aash branding is super strong and we can complement that having a strong YouTube channel. So I got many insights from the insiders going to Cosmo many cool ideas that I'm drafting in a proposal. + +Robert Del Rey: So in that proposal I'm going to show you the next phases of the Aash YouTube and how we believe we can better use this platform to educate people on a cache just to tell you a little something is that we want to focus on technical videos of a cache how to do deployments the different aspects of the console the cash chat how to contribute on GitHub those things that we say on the calls that we say them right we want to turn + +Robert Del Rey: + +Robert Del Rey: those into videos that people can see and actually learn how to get involved and contribute to a cashache. So that's the only thing I wanted to say on YouTube. at the moment we have 20 videos uploaded from the steering committee calls. I'm waiting for approval to go live. That should happen very soon. But after those videos are live, the focus is going to be on creating technical videos about the different aspects of Aash Network. Yeah, that. + +Tyler Wright: Excellent. Thank you,… + +Robert Del Rey: The next YouTube working group call, forgive me Tai, I don't know when it's going to happen. Maybe we can have it on next the 5th, or the following Tuesday. Either way works fine for me. + +Tyler Wright: Robert, for that detailed update across Zeely and the Akash YouTube. Again, there's specific working group for ZLY that meets Mondays as well as the working group around the Akash YouTube which again I will make an announcement and add it to the calendar which everybody can track at network development. So I'll put a link to that in the chat for folks to just track the calendar that can add SIGs and working groups. + +Tyler Wright: to their own calendars and stay up to date on when meetings are across the month. Go ahead, Robert. Okay. + +Robert Del Rey: + +Robert Del Rey: The third topic I had was an introduction to LATAM nodes. the guys approached me during Cosmo and they told me they have a big data center and they want to talk to the cache team and find the best ways to work together. So I don't know if that can happen by the end of this call after everybody shares their updates but just wanted to give you the heads up about latam notes. Actually, many validators passed by the cash booth saying they have resourc and also inquiring about the foundation delegation program. + + +### 00:15:00 + +Tyler Wright: + +Robert Del Rey: So, I don't know if they're in this call already, but just giving you the heads ups. Fate again. + +Tyler Wright: Excellent. I think we can take that conversation outside of the steering committee call just… + +Tyler Wright: because again we can handle that elsewhere but I'll follow up with you and get the necessary folks together to talk about that. Go ahead. Sorry if I'm butchering your name. I don't know if you're trying to speak, but I do want to move on just for the sake of time because I know there's some discussions that we want to get to as well. again, as Amanda alluded to, there's a couple of hackathons on the horizon. for folks that want to get involved or want to learn more, again, there's a SIG community where we talk about everything from events to hackathons to just the Kosh community in general. + +Tyler Wright: So again, this is where these topics are discussed in more depth. and there's more focus discussion from folks that are very passionate about these topics. So again, I don't know if Adam specifically had anything that he wanted to cover. Don't feel obliged, but again, I know that there some recent messaging that went out around the hackathon 3. We just discussed it on a couple of previous calls. but just wanted to call that out here today as another initiative as well as I believe there's a meta hackathon that's going to be going on in Brazil that a clash is going to be supporting and sponsoring in a number of different ways and that's coming up quite soon as well. + +Tyler Wright: I did want to see if there's anything else on this marketing board or anything else that we've discussed in SIGs and working groups that folks want to talk about right now before I quickly jump over to the product and engineering roadmap. Yep, we'll get to the PIP during the discussion portion. Thank you, Zeke, for bringing that up. All right, moving on to the product and engineering roadmap. again, this is another area that gets updated consistently by members of the core engineering, etc. there are various people working on projects across product and engineering side. so again, we look to get updates from folks. + +Tyler Wright: I know one of the bigger things that's been tracked is the upgrade to Cosmos SDK47. Again, we talked about this during the SIG chain. I do want to kick it over to Archer real quick just to talk about where we are. I know that we have an active test net and again some messaging is going to go out to some folks in the community that have showed interest in getting involved in that initial test net. But I just want to see if there's anything at a high level that Archer wanted to discuss around Cosmos SDK 47 upgrade or anything else on the product and engineering roadmap. + +Artur Troian: Thanks, T. I'll try to be quite quick regarding SDK47, yes, we finally started our first test net literally yesterday. there have been couple mishaps that we had with the ramping it up, but they all been sorted and that's pretty much good news. We're going to keep this test net at this moment quite low on participation. Reason being is there are still couple things that are not quite stable and we just want to make sure that we solve all of them first before we allow wider participation from the test net and our hope is that probably in two weeks we should be able to expand participation. + +Artur Troian: then moving on to the provider there have been couple things related to the provider stability things as well as the inventory like issue 259 and the core team which is diligently working to cover all of those issues and keep releasing new features as well as bug fixes. one more thing in sort of more broadband we have finally got our snap food service up and running huge thanks to Zeke for completing that task and it's pretty much done by the community. + +Artur Troian: This is great example of community participation in Akash efforts as well as a little bit of credit goes to Luna for working on our inventory exporter for the Promeigious. I saw first just literally a couple minutes ago and hopefully we're gonna get released it soon. And I think that covered it. Back to you. + + +### 00:20:00 + +Tyler Wright: Thank you, Archer. we discussed it in SIG providers yesterday. I know there's been a significant amount of work from the product team around the provider side of Akos console as a part of the console 2.0 effort. so again whether it be the free trials that have come recently, fiat payments, there's just a number of new work coming out of the console 2.0 know efforts and I think there's going to be continued work from some of the efforts from home and HQ and just continuing to again improve on the console experience. So big shout out to the product team for all the work that they've been doing and for the community for continuing testing feedback etc. Just want to see if there was anything else more specifically. Again you can get a demo. + +Tyler Wright: we got a demo on the provider side of the console from Jigger yesterday during SIG providers. The video and the transcript will be available soon. So you can take a look at that. But I just wanted to see if there was anything else specifically on the product and engineering roadmap that anybody wanted to discuss more specifically before we jump into some discussion if not then again we're at the part of the agenda where we will dive into some of the active discussions. again usually we try to carve out some time during the steering committee for discussions that are getting a fair amount of traction. I know there's two discussions that we'll be talking about today. + +Tyler Wright: One is the discussion for rare protocol and aash network full integration and the other discussion that we'll discuss in a little bit more detail is around the provider incentives pilot number two which is currently on chain for vote after again discussion for a number of weeks. I think we'll start off with the rare protocol. and then we'll take a couple of minutes to I know that there's a couple of participants that have taken some time out of their day to join us today. so I want to kick it over to Garrett and Ed from the rare protocol team. Again usually what we do is folks have already read through the discussion in great detail. + +Tyler Wright: + +Tyler Wright: Garrett and Ed, if you can give a high level on, rare protocol, the benefits the network and then maybe a followup to any of the open discussions, then we can just spend maybe two to five minutes talking about this discussion and figure out next steps and then from there we can jump into the PIP 02 discussion following that. + +Garrett Minks: Awesome. Thanks,… + +Tyler Wright: With that said, I'll hand it over to Garrett and you can just kick us off. + +Garrett Minks: And thanks so much for having us on. We've really enjoyed getting to know the Akos community and also get our open deployment layer. I would say an MVP version of our stack is already kind of up and operational on the Akos marketplace, which is super cool. and thanks so much to Zeke especially for helping us figure out exactly how to structure the SDL to get all of that working. We have our six core microservices that encapsulate the protocol are now on and kind of an MDP version on the Apache marketplace to deploy which is super cool. + +Garrett Minks: So, as a really high level, I'm the founder of Ver Protocol, also the CTO, and then Ed, who's on the call, is our CEO, and does all of our awesome partnerships and, integrations, all the business side, of the house. So, I can speak, some to the tech side and our, big picture reasoning for really wanting to decentralize the compute, layer of our stack. So we originally started as enterprise SAS dab development software products where kind of doing the same approach as the third webs and the buttons and those kind of companies of the world and through the bare market kind of decided to do this quite interesting pivot to a fully open source deployment layer. So sorry about the airport some nonsense in the background. + +Garrett Minks: so since we've kind of pivoted to open source we've had all sorts of awesome things happen to the project got picked up by quite prestigious ac So part of the Redbeard Ventures token accelerator now and has really helped us grow our dev community. we're about 3,500 devs strong in our GitHub now. And as part of this mission to kind of democratize deploying DAPs more easily and in an open source way, we also want to basically open up the compute layer and kind of get off of GCP. + + +### 00:25:00 + +Garrett Minks: when we originally developed the core software, everything was very much built around a GCP Kubernetes setup and really to scale and to fulfill the mission of having an open deployment layer, we want to have of course the underlying compute, also be open and permissionless. So, that's kind of the big goal behind what we want to do. And then our proposal is basically focusing in the first phase all of the very specific things that we think we can do to make it easier for new developers to pick up our open deployment layer and deploy it onto AOS things like kind of ironing out some of the finer points with SSL kind of see some of the other items and then with the second phase to really figure out how we can make a feature parody version of our deployment layer with all the + +Garrett Minks: scaling features and functionalities that we currently have on GCP. So, that's a high level. I don't know, Ed, if you wanted to say anything. + +Ed Prado: Yeah, I'll just add it's great to be here. We've loved working with you. We've been to several conferences. We're getting a lot of signal on Akos and certainly One of the things that was interesting when we spoke to Chang is that we represent a way to monetize and leverage your plain vanilla compute away from AI, which I think is a really important part of your model. and we also represent, a web three sort of integration and developer group that we can bring in and actually build DAPs. And as we go down this journey, there's going to be more and more integration between AI and web 3. we were just at a conference and tons of people were talking about that. So, we think that the time is right and we think we have the right product, we have the right developers, and we're excited to discuss and see if we can get this done with you guys. So, thank you for your time. + +Tyler Wright: Thank And again, I've dropped in the discussion in the chat for folks. Again, the ask is from the community pool for again their efforts. and so again, you can check out the entire ask and just get more information on the specifics on top of what Garrett and Ed just mentioned. So, is there any specific questions that anybody has at this point while we have few members from the team here live that they want to get answered? Go ahead and kneel. + +Anil Murty: + +Anil Murty: Hey guys. I'm sorry. I was just actually reviewing the proposal right now as you guys were talking. maybe I'm missing it, but what exactly is the service? If you could just explain it in one sentence or… + +Anil Murty: two sentences. What does the service do? + +Garrett Minks: Sure. + +Garrett Minks: So in a nutshell, the open deployment layer makes it easier for a new developer coming in to deploy a DAP. So instead of starting from scratch, they can start from a 95 99, percent complete DAP. So the full front end, backend, Mongo database, Reddus, all of those microservices that we have in the template that I just dropped into the chat, those are all out of the box working. And then effectively to make the DAP their own, they would go in and switch out a few of the underlying API keys. So we provide a default alchemy key say to do the syncing. they could go in modify the Docker image and then have their own syncing engine. + +Garrett Minks: And importantly, they have their own backends, too. That's something that not a lot of other companies kind of in this similar vertical to ours will open source. third web, things like that. they keep a lot of that stuff closed. so, yeah, in a nutshell, it's a open source deployment layer that makes building new distributed applications easier. + +Anil Murty: + +Anil Murty: Got it. so I guess I'm a little confused as to why. So the whole idea with console is to be the tool that people use to deploy stuff onto And in the past when we have had people talk about building potential other clients, it's been one of two ways… + + +### 00:30:00 + +Garrett Minks: It wants dark. + +Anil Murty: which is if you have other platforms that are building something for web3 or decentralized applications but trying to compete with something like a versel there's a few projects that have done that. + +Anil Murty: they'll typically integrate Akash as part of their platform and use it purely for infrastructure along with something like Filecoin or something else for storage. and that's typically something that they just do as part of their platform. what I'm not understanding here is why is this a community proposal for Akash specifically and also also just looking at the screen that Tyler is sharing that seems like a really big team for what you're talking about which is a few SDLs I mean we don't have a huge Q automation testing team just so + +Garrett Minks: Sure certainly. So I can speak to kind of So of course our infrastructure is all kind of Docker based already. So the way that we've architected it is a bunch of these containerized microservices just right now we're deploying them largely inside of Google and then our containers are being kind of created and destroyed and the GKE setup. so we have the way that we've constructed our system I think lends itself quite well to how AOS works. It's just that key difference and that key differentiator is, we don't want to have devs to have to pay fiat and use some centralized server that they could get turned off or censored, right? We want to have a native web three compute option for deploying web3 applications. + +Garrett Minks: that's the real impetus for us trying to basically fully decentralize our stack from kind of the top down because I think it kind of doesn't matter if you're using the blockchain and have all these awesome decentralized features in your application if it's ultimately running on a compute layer that is run by Google. so yeah, that's the, high level reason why we want to do it. And then the more sort of brass tax way that we're able to accomplish it is by basically porting over all of our existing config maps like existing manifest files, all that stuff that we've done in Google. + +Garrett Minks: we're basically migrating and moving over and it seems like Jiao asked about the yncing. So we have two ways to do syncing in our system. one is a kind of out of the box solution that uses alchemy largely to do all the RPC type services. We also have a native one where you can pop in your own RPC URLs into the environment variable file of the docker at build time and then basically sync that way and then all of those transactions when they sync will then get stored in the mongo which is another microser that gets deployed nicely when that one SDL is run. + +Garrett Minks: So basically when somebody deploys a DAP with our system they're running their own entire self-s sovereign self-contained DAP infrastructure including the backend in the rare node microser and… + +Tyler Wright: I know we need some time for other discussions. + +Garrett Minks: then their own mongo in that micros service that helps + +Tyler Wright: I would just ask that anybody here today if they haven't already, please feel free to drop comments and discussion. Again, this is how we continue to make sure that the community is as involved with everything that has to do with community pool spend. if anybody has any additional questions, thoughts, concerns, votes of approval, whatever the case may be, ple please drop those in the GitHub discussion. and then we'll continue to again fine-tune and figure out what the next steps are. But very appreciative of Garrett and Ed taking the time out of their day to join us today and safe travels, Garrett, for sure. But thank you very much for taking the time. + +Garrett Minks: Awesome so much. Thanks for having us and I'll hang on for the rest of the call too. + +Tyler Wright: + +Garrett Minks: Okay. + +Tyler Wright: Excellent. All right. + +Tyler Wright: Continuing to move along. again, there's a couple of other items that, are on chain. so again, please get involved in some of these discussions. I know that Huey dropped in some, metrics from Q3 events in some of the past quarters. So, please track that. but right now I do want to because I know that folks have been talking about it and want to talk about a little bit more. I do want to kick it over to Chang to talk a little bit more at a high level. I know we've talked about it in previous calls prior to this discussion going up but talk about PIP at the highest level and… + + +### 00:35:00 + +Tyler Wright: then see if there's any questions from anybody here that can be discussed. + +Cheng Wang: Thank you,… + +Cheng Wang: Ty And let me know if Changbot v1.2 is bothering you guys and I'll remove the filter here. so in terms of PIP provider incentives pilot number two, this is the follow on to provider incentives number one, right? the first level of incentives was just for a quick recap in everyone's information was approximately $5 million in size and funded approximately 400 additional GPUs that have all today to this day been added to AOS network. And as far as where we are in the commitment of those funds, initially we started I think the first contract with a provider or two back in March of this year and throughout between March and this point in time we are still in the process of fully deploying All those funds at this point in time have been fully committed. Of course the spend will continue to move forward. + +Cheng Wang: So we're at peak basically GPU capacity from pip 01 chips at this moment in time and pip 02 is a double of that size up to 10 million and there are a lot of community discussion around that particular item in GitHub as well and so currently this I think is already on chain for vote and the idea here if you can scroll down a little bit so those chips up top in the current table that's been displayed right now are the ones that were committed to in PIB01. And in the table a little bit further down, these ones right here are the ones that we're expected to, we're estimated to deploy funds towards for PIP 02. Again, these are, high level est estimates based on the best of our understanding with us and the core team rather in the community at large, And this mix can change a little bit over time, right? So this is the essential makeup here. + +Cheng Wang: the biggest thing I'll say is the idea here is to bring greater numbers of highly sought-after GPUs onto a Kosh network to be able to power larger workloads with larger tenants, right? those are all extremely important especially as some of these things begin to scale. We're already running up against maximum utilization as you guys will see if you go to the stats network today. with trial wallets as well as with credit card payments up online now we're essentially red line on all the high-end GPUs we have and we're at near I think record usage today which is incredible from GPU side of things so if there are any questions please drop them in 810080s of course the most sought actor models there and L40s etc will be kind of the replacement model for the A180s MI300's + +Cheng Wang: and the H200's are the looking forward set of models, right? kind of the newer generation and of course the RTX490 is crowd favorite there as far as inference capabilities go in a distributed fashion. and A6000s which kind of round out the lower end of those particular models to service clients on that side of the spectrum. That's Fire away. + +Tyler Wright: Perfect. Go ahead, Zeke. + +Zeke Ezagui: + +Zeke Ezagui: So, a couple questions. I'll start out with, it looks like the timeline for this is going to be 6 to9 months. Is there any thought in wanting to do 5090s … + +Cheng Wang: + +Zeke Ezagui: because those are going to release in spring of + +Cheng Wang: That's a great question. + +Cheng Wang: Probably I think just to give you an idea in terms of committing to these ships. So, hypothetically we go out to six, eight different providers and look for these resources. right 4090s we have 204 here kind of roughly allocated. That doesn't mean we'll take down a swad of 204 all at once. It'll probably be done in tan so in that vein, if there's, excess funds and we want to reallocate when the 5090s come out, I'm sure, we'll do so. I currently not envisioned here, but I think we should be flexible over time for But Neil, I was just going to say if you could please take that on, that'd be great. + +Anil Murty: + +Anil Murty: Yeah, and that's a good question, one of the other things to keep in mind here, and Chen, keep me honest on these numbers because we did them a while back. I may be, confusing some of them. the numbers that are on this chart that Tyler is sharing, these include GPUs that are committed currently and GPUs that will be added in the future… + +Cheng Wang: That's + +Anil Murty: because as Chang said, if you go and look at network today pretty much all the GPUs of all the models are being utilized and so our utilization rate is really high. What that means is that there are actual customer workloads running on the GPUs that we have committed contracts to right now. And so what that means then is that as these contracts get to their end, we can't just abandon the existing GPUs that have customer workloads running on them. So we have one of two options. + + +### 00:40:00 + +Anil Murty: we can either work with those existing customers and offer them potentially the next generation GPU and then transition them to these new clusters or we can continue maintaining the existing clusters. So I think we'd have to sort of look at these a little bit on a case by case basis. but your point is completely valid which we have to look at the next generation of GPUs in many of these cases. + +Tyler Wright: Excellent. Thank you. + +Zeke Ezagui: Yeah, that makes sense. and my next question is there was added after some discussion the 5 to 10% for community members. what's going to be kind of the process on if I'm a community member, how do I go about trying to get, approval or whatever getting some incentives to do so? what's going to be the process for + +Cheng Wang: + +Cheng Wang: I think we'll have to probably iron that out when the time comes. It really depends on I think the parameters that we set or are roughly that of a tier 2 data center requirements, right? So, with that being said, I think as long as the u community provider can uphold these, of course, quality, availability, density, all those requirements, and I think it's really just an open discussion. I don't think there's anything too structured at this point in time. Anal, I hate to hot potato this over to you once again, but I'd figure it seems pretty straightforward, the structure here or the requirements. + +Anil Murty: + +Tyler Wright: Go ahead and + +Cheng Wang: So, I don't imagine there's too much vagaries here to deal with. + +Anil Murty: Yeah,… + +Anil Murty: I think yeah, so I think Tyler is sharing that portion of the screen, Zeke. after some of the feedback from you and other folks that have been really strong providers on the network for a long time, we added those bullet points specifically for trying to outline what would be potentially the criteria for somebody that could be eligible for the community portion of this pool including the percentage allocation to community providers. + +Anil Murty: So, definitely take a look at that if you haven't already and please provide feedback if that sounds good to you. But that's a general idea. and again, this is kind of a precursor to us implementing an onchain mechanism for allocating funds to all community providers and not just a subset of them. + +Zeke Ezagui: Yeah, makes sense. So that just be, a GitHub discussion. + +Zeke Ezagui: That would be messaging someone at the core team. that's kind of what I was talking about. + +Anil Murty: That's correct. + +Anil Murty: Yeah, it'll be pretty manual for now because we just don't have the resources within the core team to go and implement what is needed for us to be able to have onchain incentives and not get into some of the problems that some of the other GPU marketplaces have had with civil attacks and… + +Tyler Wright: Wait, you + +Anil Murty: and spam accounts and things like that. + +Anil Murty: + +Zeke Ezagui: And… + +Zeke Ezagui: if I'm interested, would that just be, sending you or chain a message? Would that be sending Tyler message? Would that be, doing the GitHub discussion? + +Anil Murty: yeah so I think… + +Anil Murty: if there are people that are within the community that meet the requirements of so I guess there's two steps to this right one is the community and people that review this proposal agreeing on the criteria that's outlined in the proposal here specifically the bullet points below the + +Anil Murty: + +Anil Murty: that Tyler is sharing. and then assuming that that is good and this proposal actually passes then the pro process would be if the community provider meets those requirements then the core team being the stewards of driving the provider incentives pilot would ensure that those criteria are met and… + +Anil Murty: then ensure that there is fair allocation of resources across all the providers that meet That criteria. + +Tyler Wright: much appreciate the questions and… + +Tyler Wright: the responses. Again, if anybody has anything that any other questions, thoughts, feel free to drop in the GitHub discussion. I'll follow up with any updates as pertains to just structure. again, as Anil mentioned, it's going to be manual, but we can fine-tune where the best place for people to interact. and I'll make that a bit I'll drop a comment over the next couple of days with how to do that. Again, this is on chain for vote right now. + + +### 00:45:00 + +Tyler Wright: so again please vote if you haven't and again continue to drop in messages in this discussion and we'll continue to track it at sick rs or Any other questions as it pertains to PIP number two? Terrific. I'll also drop in the recording and transcript from this steering committee meeting as a comment as well just so people have it easily Real quick another discussion update that I want to call out Adam with support from other members of the team has updated best practices for governance proposals. + +Tyler Wright: Again, this governance proposal is the best way to get funds from the community pool. Again, the community pool is available to all members of the community and is a great resource to get people building and helping to grow the Akash network through feedback, through the last year plus of seeing things in action. Again, there have been some updates made to the governance proposal guidelines. So, please take a look at those and it'll tell you, how to start off, from an idea to getting a request from the community pool and having conversations with the community. So, again, if anyone has any questions, feel free to drop them as comments, but play please take a look at that discussion around best practices for a kosh governance proposals. All right. + +Tyler Wright: Is there any other discussions that anyone wants to talk about right now? All right. If not, then usually this is a portion where we just open it up for discussion. real quick, I know we only have nine minutes left. I want to be respectful of folks time because I know people take time out of their busy days across both the core team as well as other community members. So is there any other topics that anybody wants to discuss right now? much appreciate rare protocol team for joining us. thank you to Chang and Anneil for fielding questions around PIP02. again please get involved in discussions. + +Tyler Wright: Please stay up todate on all the things that are going on across the Akos community by looking at the project boards. There's a number of ways to contribute if you look at the bounty boards. As you look ahead again, there's a working group for Akos YouTube that will be added to the calendar soon. ZLE meetings every week. there's also going to be, SIG chain where we'll talk a little bit more about the Cosmos SDK 47 upgrade. again we get updates on all things provider related at sig providers. So there's just a number of meetings that happen throughout the month where you can get involved have very very specific discussions around themes and topics. I know there's a number of folks that are going to get more involved in SIG support. Again support is where issues that are related to the AOS blockchain and just a core codebase go. + +Tyler Wright: + +Tyler Wright: So again, if you want to get involved and contribute there or just learn more about how Akos works, under the hood, please get involved in sick support. Much appreciate everyone's time today. Look out for announcements on discord on social media and again just continue to just reach out and stay involved. But again, much appreciate everyone's time today. + +Tyler Wright: Hope everyone has a great rest of their day and… + +Robert Del Rey: Thank you,… + +Robert Del Rey: Tyler. Stay safe, everybody. + +Tyler Wright: a happy Halloween if you Be in the metaverse. + +Zeke Ezagui: Thanks everybody. + +Cheng Wang: Thank you Tyler. Take care. + + +### Meeting ended after 00:49:38 👋 +