Chainflip Community Update #6

Chainflip Community Update #6

Hello Friends!

What a time the world is having! Between war, inflation, destabilising global economic shenanigans, and a general sense of disruption and chaos in the air, the last 4 months or so have brought us a rather different situation in the Web3 world compared to what we’ve experienced over the last 2 years or so.

Stable Diffusion generated art with the prompt: "2022"

Yet, despite this, the show goes on, and things have held up better than one might expect with such strong headwinds. The Web3 space is proving its maturity and resilience. To me, what we’re all going through now feels completely different to 2018 and 2019 where things went dead silent. In fact, the last quarter has presented some unique opportunities for projects in a position to take advantage of an increased talent pool from layoffs across the tech space, a less demanding market environment, and the time and space to forge ahead and make progress.

I’m very pleased to report that Chainflip absolutely fits into that category. When we started building Chainflip in 2020, we didn’t expect to still be pre-product and pre-token at this point in time, but it is what it is, and the lessons learned over the last two years have helped us evolve the project beyond its initial concept and scope. I apologise for the quieter-than usual quarter from us, but we’ve been real busy, and have done a LOT in the engine room over the last quarter, and as such, we’ve got a rather exciting set of updates to bring you up to speed on today.

However, like everyone these days, we have experienced some turbulence, but I believe that we’ve emerged stronger on the other side, so I’ll tell you a couple of stories about that too.

And of course, the million dollar question: “Wen Chainflip?” - I’ll give you an updated answer with enthusiasm.

Summary of Updates:

  1. Sandstorm is READY!
  2. Ibiza UNDERWAY!
  3. Perseverance LAUNCH!
  4. INBOUND: New Website, Lightpaper, and MORE!
  5. JIT AMM python model DONE!
  6. Announcing JIT Software Strategies…HUGE!
  7. Berlin Blockchain Week: MERGE NACHT!
  8. The C to the T to the O!
  9. New roles and team GROWTH!
  10. Aggregate or DIE!
  11. What’s next and WEN!

Numero Uno: Sandstorm is READY!

Now that I’ve got your attention, let me explain.

Sandstorm is feature complete. That means all of the bits and pieces needed to say that it’s fully functional are in the codebase and working. However, there is still a decent amount of polishing, refining and double checking to be done. We think that Perseverance will be the key tool to ensure that we achieve this, which I’ll get into in a second.

Over the last quarter, we’ve put in tokenholder governance, added a lot of smaller features to make sure that the network can handle itself more gracefully, and critically, we’ve performed triple-bypass surgery on the communications layer within Chainflip Validators that allows those in the authority set to directly connect to each other over ZeroMQ for all key-management related stuff, including KeyGen and Signing ceremonies for TSS. It took a couple of months to develop, and we finally tested the hypothesis in a nail-biting deployment last week.

The results? GOOD. In our testing of the new communications layer (and other fixes and upgrades accumulated since August), we successfully and repeatedly completed full authority set rotations with 150 validators in a very consistent 6 minutes with absolutely zero attrition of nodes. That’s auction resolution, keygen ceremony, witness confirmation, a signing ceremony to verify/test the key and witnessing the results of that, Ethereum key rotation, rotation witnessing, and finally transferring to the new authority set, all happening reliably and gracefully. Compared to where we were in Paradise Net, where we would be lucky to get 147 nodes after many unsuccessful runs over an 8 hour long rotation period, it is simply leaps and bounds better in every measurable way.

However, I’ve got to set expectations here: this was on an internal network split across a 3-region, 3-cluster AWS network, and as our Testnet programs have demonstrated, sometimes things don’t go to plan when the validator software is put into the wild. We’ll come back to the Perseverance program shortly, but assuming that it’s immediately successful, what would it mean?

Does it mean we are LBP/Launch ready?

Yes and no. Mainnet launch is no joke. We only get one shot at it, and there’s three reasons why pulling the trigger right now would be less than wise:

  1. Polishing and refining the Sandstorm features will make a big difference for scalability and security for the exchange itself, and there is a known list of things to check off and test in the wild to satisfy our stringent confidence levels;
  2. We need to fill some new roles like Support Engineer and Ecosystem Development to be able to handle all of the new overhead that comes with hitting mainnet. More on that later in the update;
  3. And most important: we need to catch-up with the AMM/DEX features before doing the mainnet launch. Launching too early will only slow us down after that point, and we don’t want to have only the Sandstorm features on mainnet for months before making swapping go live. A token and network is good, but a token and a network with a product is the only thing that will make it count.

Luckily, there’s news on the Ibiza swapping release too:

Number Two: Ibiza is UNDERWAY!

Broadly, our development plan has always been to develop the Chainflip experience in stages, starting with the underlying network that makes up the Sandstorm feature list. In the last 2 months, we have shifted well over half of our development resources to tackling the Ibiza feature set while we clean up stuff like the new ZMQ system and other web and infrastructure work.

Ibiza can kind of be thought of as Swapping Lite™.

It’s got the fully featured JIT AMM, it can ingress and egress ERC20 tokens, native ETH, substrate assets (starting with DOT), swap USDC for whatever, and so on. It doesn’t have fully featured asset routing, and it won’t have to prepare network and relayer fees and so on, but it is the minimum viable product that does the bulk of the heavy lifting on the protocol side. Basically it’s all of the cross-chain AMM “need to haves” rather than the “should haves.”

The team has been making strong progress on this, which you might have seen if you’ve been following the development updates. We can now ingress Ethereum and ERC20 tokens quite happily. We’re working on broadcasting them, and we’re yet to implement the JIT AMM on the state chain directly, but with the exception of the AMM code, we’re reusing a lot of stuff that we already did in Sandstorm, so really this isn’t too much of a lift.

Our internal goal is to have us ingesting and then spitting out whatever assets you want in and out of our vaults by October 31st, which so far looks like we’ll beat that target. Once we have done that, all we gotta do (oversimplifying, of course) is put an AMM in the middle and we have ourselves an Ibiza - and the AMM model is looking slick. More on that later. Once Ibiza is done, it will be pushed to testnet for all to see.

Provided we keep up this pace, hitting our Trail of Bits audit deadline of January looks achievable, and from there, we are good to go. Let’s hope nothing explodes in our face between now and then.

While we’re here, if anyone would be interested in helping us write an Ibiza integration for Bitcoin and other Bitcoin-based Schnorr compatible assets, hit us up.

Berghain, the third and final stage in our bootstrapping process, polishes off the work on top of Ibiza and adds cross-chain aggregation, deeper integrations, better gas fee handling, and extra stuff like the buying and burning of FLIP. That is the product we want to ship to mainnet, but to get there we need a strong backbone, and Ibiza is the Beta that confirms that backbone for us. If Chainflip were a Reuben Sandwich, Sandstorm is the rye bread, Ibiza is the pastrami, and Berghain is the pickle on top. But who wants a sandwich without a f*kin pickle, eh?

Numer Trzy: Perseverance Persistent Testnet LAUNCH!

“What kinda name is that?”, you ask. Well depending on who you ask, you might get one of three answers. The first might say “Mars rovers are pretty cool, yo.” The next might realise that it is actually the name of an old pub/club in Melbourne’s northern suburbs which has been home to all of the worst 18 year olds in town (including myself, once upon a time). The third knows that the common link between the two is that they both refuse to die - and that is what sets Perseverance apart.

Perseverance will not be retired like our previous generation of testnets. It will live on indefinitely, being upgraded as new features are released, and will help us maintain a production standard even after mainnet is released thanks to the amazing help of the community that have formed around Chainflip. It’s our persistent testnet that will live alongside Mainnet: one we care about and will try to keep alive to ensure we are prepared for all scenarios that we may also see on Mainnet. It will also exist alongside any temporary networks we create to quickly test and break new features in a faster environment.

Rather than talk too much about it, we’ve got a blog post hitting the shelves very shortly that will go into MUCH more detail about the program - all you need to know is that Perseverance goes live on October 12th, on time, and as scheduled (hopefully). Let’s hope the p2p gains seen in testing are reflected in this production-like environment!

Número Catro: INBOUND: New Website, Lightpaper, and MORE!

Yep. Times are changing and so too must the brand. Expect an overhaul of our general first-impression appearances in the next little while. This is in preparation for the upcoming wave of launches from us that hopefully you’re getting a sense for!

Nummer Fem: JIT AMM Model is DONE!

The JIT AMM is not just Uniswap v3 done in substrate…oh no! It’s an advanced hybrid of range orders and maker-only limit orders all wrapped into one system designed to blend the best of both worlds by offering liquidity providers complete flexibility in their market making approach on a per-pair basis. No fragmentation, no dilution, and fair distribution of liquidity fees for all participants.

We think it will be the most capital efficient AMM system in existence. It’s also pretty efficient from a code perspective too, thanks to the amazing work of Albert (DJ Tinhat), Alastair (Flippant Nickel Bag), Dr. Loveless (That’s actually his name), and several others, who have worked tirelessly to produce a Python copy of Uniswap v3, and then prototyped several ideas to arrive at a final solution implemented on this model much faster than would have been the case had we tried to do this directly on the state chain.

We’ll be open-sourcing both Python models shortly for anyone wanting to experiment or backtest using either of these AMM models. In particular, the models are proving to be very useful for our sister company’s task.

Uimhir a Sé: Announcing JIT Software Strategies…HUGE!

You will probably be hearing a bit more about JIT Strategies over the coming months and years. JIT Strategies is our Dublin-based sister company set up to create and manage the Chainflip organisation’s Market Making Desk and Production Validator Operations.

We’ll let Dr. Loveless, the merciless and cold-hearted leader of the team, do the talking when he’s ready, but I’m personally very excited to have a new firm within the group taking care of some very important parts of business so that we at Chainflip Labs can focus on our core task of Protocol & Ecosystem Development.

Numéro Cinq: The C to the T to the O!

This one is a fairly big one. Lots of you probably saw, met, or heard the bearded Tom Nash over the last couple of years at Chainflip. Some eagle eyed community members have noticed that he hasn’t been noticed around the usual spots and have started to wonder what’s going on.

It’s with a heavy heart that I have to confirm the rumours. Tom has indeed had to leave the Chainflip team after a solid 18 months of grinding, tenacity, and dev updates, for personal reasons.

Tom and I go back a long way. We met many years ago in the Melbourne blockchain scene and spent many an evening goofing around in the Flex and Loki offices. We’ve seen a lot together. We both moved to Berlin for Chainflip. Side by side, we got rained on for 3 days straight at the Belgian Grand Prix last year. And most notably, we have each tried our very best to steer the Chainflip vessel through all sorts of weather. While I won’t lie and say that it's always easy to start working with your mates, I want to put on record my thanks to Tom for all that he’s done to help get us to where we are today. We will all miss him (and that beard) and wish him all the very best in what lies ahead for him.

In the interim period after Tom’s departure, we’ve implemented new internal structures and I’m pleased to say that we think they are working well. Dan (MC Method Machine) - who you may have seen has largely taken on the Dev Update mantle - now leads the Rust/Protocol team, Tom J (DJ Govner) leads the Intra/Platform unit, and Andrew (DJ Dibble) has stepped up to Web Lead.

Chainflip does need a CTO. Once we go live, there just won’t be a way to keep on top of all of the aspects of protocol development without a capable pair of hands running the engine room. After just a few weeks of searching for a new CTO, I was getting quite overwhelmed by the sheer volume of people offering to fill the boots of such an important position in the project.

I’m very pleased to announce that one person rose to the top of that pile. In our several interviews with him, the only negative feedback that any of us could identify about the guy was that we all thought that we’d missed something. We couldn’t think of anything bad, and that made us suspicious. If you’re reading this, new CTO, be warned. We are watching :)

I can’t tell you the name of our new guy just yet, and it’ll still be a couple of months before he will be able to join us full time, but here’s what I can reveal:

  • He has a PhD in Computational Physics (nnneeeerrrddd)
  • He worked in London for years in the finance industry building quantitative computational price risk models for Debit…..ehem…I mean CREDIT Suisse
  • His most recent engagement over the last few years has seen him scale a Web3 security-related team from one to several dozen engineers
  • He has a dog that he strokes in board meetings like a Bond villain
  • He has agreed to take me up for a flight as he’s getting his pilot's licence (posting here for the record)
  • He can also DJ (of course)

I’d say more, but for now I can’t due to some confidentiality stuff. We’ll do a proper introduction soon! For now, he will be known only as Dr. Goldfinger.

Nummer ACHT: Berlin Blockchain Week: MERGE NACHT!

As you may have seen, we put together a huge rave at Berlin Blockchain Week. We booked Aeden, one of Berlin’s many famous techno clubs thanks to a hookup through DJ Midday, and had over 1000 people apply for the guestlist of Merge Nacht.

I was quite terrified of DJing in front of several hundred people for the first time in a long time, and the first time in a club, but was pretty happy with the end result.

Sorry that we couldn’t stream it. Venue policy prevented any and all cameras from being used during the event, but here’s my set and a pretty illegal picture to give you a feel for what it was like!

We met community living legend Sleipnir who drove all the way from Poland for the party, as well as dozens and dozens of people from the Ethereum developer community, Parity, Cosmos, Solana, and many more besides. It’s all about making friends, and that was certainly done to a high standard.

Keep an eye on - we may do this again in good time. Hopefully DJ Xtroleum can get out the house to make it to the next one!

Nommer Nege: New roles and team GROWTH!

We have hired a fantastic Web3 Product Manager to help us streamline our product development processes and gather more user feedback, design better specifications, and generally speed up our processes to get us ahead of the rest. Expect an introduction to this mysterious individual in just a couple of short weeks, but this is just the start.

Support Engineer. Ecosystem Development Manager. Company Assistant. Content Designer.

These are just some of the new roles we’ve opened up in the last couple of months in preparation for a huge 2023. It’s going to be a lot of work to make Chainflip the premier on-chain market for cross-chain swaps, and we need some help doing it. These new hires will make the Chainflip launch a lot better.

If you would like to join Chainflip Labs in a role that is *not* Rust Engineer (I bet many of you are sick of hearing that one!) - this may be your opportunity!

Check out our current listings here:

номер десять: Aggregate or DIE!

We’re working on an investigation into how to best get Chainflip integrated with aggregators, and to have aggregators integrate into the Chainflip frontend to make sure that we can meet as much demand for swaps as possible, even if we don't support particular assets or DEXes out of the gate.

If we can beat other protocols on pricing for every pair that we list, and list assets that other protocols can’t, we think that this strategy will put Chainflip in front of other cross-chain products suffering from the fragmentation hangover that comes with bridging and cross-chain messaging solutions. Effective aggregation makes that competitive edge harder to compete with, and that’s really where we want to play. It also means Chainflip’s frontend will potentially be able to support truly epic numbers of assets on day one, allowing our users to stick with us as we grow our own native pools without losing out.

It also means we don’t have to bother trying to organise and develop USDC pools on 15 chains when Circle now supports this functionality natively, and we can piggyback off that. Easy. (Theoretically)

This is all about offering users the best possible experience, and fulfilling our mission of taking all crypto spot markets on-chain, and helping to make the Web3 space capable of supporting users without needing centralised exchanges.

I’ll be back in a month or so with our findings.

il Finale: What’s next and WEN!

Well damn. That was a lot. I suppose I have beaten around the bush enough, and should probably just get to the question on everyone’s mind: wen.

I hope this update has given you a lot of data on where we’re at and what's going on. Soon you can expect a lot more noise from us now that our metaphorical ducks are in a row, but as you may have gathered, we won’t be dropping Berghain until at least our security audits are complete, and with Sandstorm still needing a bit of attention from the janitor after the Perseverance launch next week.

We don’t just want to launch Chainflip. It needs not only to be good, but needs to be in a position where it can rapidly evolve once it has been launched. Give us until Christmas, and we should be in a position to put forward a date and a time, not just for the FLIP and Sandstorm mainnet drop, but also for the Ibiza Beta, Berghain launch, and a roadmap for the rest of 2023.

It’s taken longer than any of us wanted, but it will be worth it - especially with the amazing community we are so fortunate to have gathered around us.

I hope you have kept well in Q3, and that the world hasn’t gotten you down. Together, I know we can make it through. Hop in the Discord and Persevere with us!