CasperLabs Telegram AMA Recap

First of all, Mrinal, could you give us a brief intro of yourself?

Hi, I’m Mrinal, One of the cofounders of CasperLabs. I’ve been fascinated with technology from a young age and began programming at 11.

 I got my Masters degree in Computer Science from Carnegie Mellon University and after internships at Microsoft at Oracle and Microsoft went to work with Bain & Company, Bain Capital and then was the Sector head for Technology at $1.2Bn hedge fund. I’ve been in the distributed computing space for a long time, programming as an open-source contributor to BitTorrent in 2002 and then Investing in Cryptocurrency since 2012. I Invested at the earliest stages in Ethereum, Blockstack, Basis, Maker, Filecoin etc. 

I love this industry and am excited that we are just getting started. Excited to answer your questions.

In the blockchain industry and decentralized world, the "public chain" is not a new word; we have witnessed the development from public chin 1.0 (Bitcoin) to public chain 2.0 (Ethereum). Nowadays, a lot of public chains are focusing on solving the problems of system scalability, security, and regulatory compatibility. In your opinion, what issues do you think still exist in the use of the underlying public chain?

Well, I think it’s important to first note that every new technology takes a significant amount of time to gain mainstream adoption. For example, Amazon Web Service (AWS) took 9 years to cross the chasm, and the internet took decades to become the phenomenon that it has become.

The companies that are able to cross the chasm to mass adoption do a few things very well.  Firstly, they address a customer need.  All features either increase revenue or reduce risk.  Even though blockchain is a trust layer, (it reduces risk) working with public blockchains is very risky.  Here is why:

-It’s hard and expensive to find blockchain engineers, due to custom programming languages.  (Keyman risk)

-It’s hard to test your on-chain code (Defect risk). 

-It’s impossible to run tests against the on-chain code using existing CI/CD workflows - you cannot run a node on a CI/CD server.  (Delivery risk, Defect risk)

-On-Chain contracts can’t be upgraded. (Defect risk- business continuity risk)

-Access control mechanisms for accounts and contracts are not usable/complex. (Business continuity risk)

-It’s hard to budget for transaction fees in tokens. (Financial risk)

-Nodes are hard to monitor for uptime. (Business continuity risk)

We have solved ALL of these problems.  Without removing these risks to adoption, it’s not feasible to build a business using a public blockchain.  This is why we see private blockchain adoption.  Many of these risks go away.

In Addition, there are no public blockchains that have achieved Scalability while maintaining full Security (not just probabilistic security) and true Decentralization. Most of the newer blockchains have some centralizing features as well as a tendency to have very concentrated ownership structures. 

We do not compromise decentralization while having correct-by-construction security (which means the Math comes first). Every decision has been made to broaden participation, ownership and ease of use.

Mrinal, there are a LOT of public chains these days. New protocols are being announced almost every week. 

How does CasperLabs different from other public chains? Can you highlight some unique features of CasperLabs?

Well, I can specifically talk about things we do that help developer engagement and ease of use. Let’s talk about some of the specific things we are doing.

Firstly, we have embraced open programming standards. 

The chart on the left shows that while solidity is the most used smart contracting language today with ~10k developers, this number is much smaller than the 26 million total developers out there.

The chart on the right shows that Rust has the highest approval rating amongst programming languages (this is from Stack Overflow’s survey which had over 90,000 respondents) 

This tells us that for blockchain to get mass adoption, smart contracting interfaces need to be easy for all developers and they have made their preferences for 1. broadly used and 2. well-supported programming languages are very clear. We support Rust and Aseemblyscript right now and over time any programming language that compiles to WASM, hence a blockchain for all 26M developers out there,

Secondly, we have smart contracting features that make the developer experience significantly easier, such as

Flexible payment code (Sender does not always have to pay)

Weighted key management (You can create any kind of threshold system)

Upgradeable contracts (Never lose control of your systems); 

and Direct state querying (Easy data access)

The above 4 features (and many more) are unique to our deployment.

Mrinal, this looks so well-though and developed to me. A lot of blockchain projects need this. 

CasperLabs's technology is powered by a Proof-of-Stake (PoS) CBC- Casper protocol, which once brought a huge discussion in the community. Can you briefly introduce and elaborate on this protocol?

There are several differences between the Proof of Work (PoW) and Proof-of-Stake (PoS). I will layout 2 key differences and then talk about why CBC (Correct by Construction) is important.

The first key difference is the usage of computational power. In PoW 90-95% of computing power is used to generate hashes (Random numbers) and hence there is a very high hardware and energy requirement. In PoS most of this processing power can be redirected to performing actual work. Thereby increasing efficiency as well as speed of the underlying networks.

Secondly, proof-of-work encourages centralization versus proof-of-stake. The reason is in PoW your reward is based on the amount of relative processing power and is randomized. So if you have 0.5% of the processing power on PoW, you have a 1/200 chance of ever earning a block reward and it might even take days to earn anything. This is why you have very large mining operations that try to make the rewards more predictable by consolidating power. In proof of stake since rewards are not randomized (i.e., since they are split by stake), Therefore even small validators can participate meaningfully in the network.

Our protocol is unique in that it is built using the CBC (Correct by construction) methodology. This means we start with the mathematical foundations of the underlying behaviour and then start creating /coding the protocol. We are very fortunate to have excellent researchers on the team including Dr Daniel Kane (Eduard will share his Bio-Link, to avoid spam) who helped derive the mathematical proof of the underlying protocol. Dr Kane is a mathematical prodigy with degrees from MIT and Harvard (PhD)  as well as winning major mathematical prizes in the USA. 

Bio: https://en.wikipedia.org/wiki/Daniel_Kane_(mathematician)

Paper: https://github.com/CasperLabs/highway/releases/download/v1.0/highway.pdf

This work as shown in our mathematical research paper Eduard will share the link) has allowed us to produce a protocol that gives full security, does not compromise decentralization and is mathematically derived so it is not just probabilistic.  We will have this peer-reviewed before launch.

Mrinal, we all know about your successful Series A. It was really impressive. But what progress did you make since then?

It has been a very busy period since we closed our Series A. Our focus has been building the blockchain and the team.  The technology team is now fully built out with 22+ researchers and engineers. We release updates every month and are progressing according to plan, including our testnet launch on March 31st. We launched our testnet in collaboration with validators from some of the largest staking pools in the industry, such as SNZ, Stake. Fish and Hashquark.  

You can interact with the testnet at our website (https://clarity.casperlabs.io )

Below are our roadmap and we look forward to launching the main net by the end of the year.

In terms of blockchain centric partnerships. We have made a lot of progress with staking pools and leading exchanges. We will have dozens of staking pools and key exchanges acting as initial validators on the network. We believe this will greatly help create a decentralized and distributed partnership base.

Also, we have been able to fine-tune the protocol and feature set through the excellent relationships with both blockchain and non-blockchain projects that our investor and advisor pool brings. A lot of the core decisions made on the technology are direct feedback from some of the most innovative companies out there, combined with the professional experience from our senior engineers on the team.

Now, let's please focus on CLX token. What are the use cases?

CLX functions just like eth on the Ethereum network. It pays for transaction processing. It addition it's also the staking token.

So a very simple, flexible token design. It can be used to pay for any kind of computation and use case. People can also create their own tokens similar to the ERC-20 standard and use CLX to pay for the underlying transfer/gas fees

What is the design logic for the token to be issued?

For some context, we have rigorously avoided any real concentration of ownership. This is the reason why when we raised $14.5m in our series A we did not sell a single token; and also raised that capital across 60+ contributors even at the earliest stages.

As a result, the high-level breakdown of tokens is as follows...

1) The team only keeps ONLY 8% of tokens. Given this is shared by 40+ individuals there is the very limited concentration at the individual level and it releases on a 3-year vesting schedule

2) We have allocated 16% to developer and community incentives which will be governed by a Developer DAO. We already have luminaries from companies such as AMAZON, Microsoft working on this DAO and will have some announcements around this as we get closer to launch.  This will be a combination of a credits system similar to Amazon AWS to help companies build on the network as well as incentives for developers to help upgrade and augment the system. The idea is to deploy this over 5 years. This tranche will greatly help accelerate the adoption of the underlying blockchain

3) We have allocated 6% for advisors and strategic partners. This will work similarly to developer and community incentives and will be distributed over the long term over multiple years. 

4) Only 10% will be held by the CasperLabs network; any and all distributions will be commensurate with public release, hence very long term. 

5) The remaining 60% will be distributed to the public via our validator sale as well as ongoing public sales.

Will CasperLabs produce any DeFi products in the future? This industry is pumping right now Will CasperLabs produce any DeFi products in the future? This industry is pumping right now.

We are an infrastructure layer but will partner with DeFi products. We actually have a few features that make DeFi pretty easy on our platform. First of all Flexible payment code allows a much simpler DeFi experience on our system. Example below:

In addition, we support weighted key management which is extremely important to DeFi products like structure credit as well as Equity type votes, the description below.

How can CasperLabs solutions help with predicting the costs of an app?

The protocol will Flex the gas rate based on the Fiat/CLX pair. We have a partnership with Chainlink and have already begun implementing this feature. The diagram below shows how it works. But in essence, a set of decentralized oracles will help verify feed accuracy.

How did the partnership with Hyundai go?

They were a significant Series A investor and the partnership is still strong. In fact, we had a conference call with them yesterday. They have licensed our execution engine for their own deployment. We are working on some really interesting far-reaching use cases with them, announcements will be made as these firm up

Could you tell more about the team of CasperLabs, who doest it consists of?

We have 30 full-time employees, over 20 are developers and researchers

In addition to Dr Daniel Kane whom I mentioned earlier.

Dr Andreas Fackler got his PhD at Ludwig-Maximilians-Universität. He held roles at Google for several years and has also worked on the Maidsafe and Honeybadger protocols.

Medha Parlikar is our CTO and one of our founders. Medha comes from industry -delivering production software that has to be up and service thousands of requests per second in mission-critical applications.

Cliff Sarkin, our COO, is a tech entrepreneur and licensed attorney. He has worked with several startups including leading VideoSurf’s $80M acquisition by Microsoft in 2011. He holds a BA from the University of California Berkeley and JD from Harvard Law School.

I wish I had time to talk about everyone at the company, but everyone has similar backgrounds. For example, over 80% of the team have Masters and graduate degrees most from Ivy League or equivalent schools. On average members of the team have about 15 years of experience, both at leading companies like Bain Capital, Microsoft, Google, Adobe, Amazon, DropBox, but also protocols such as Ethereum, HoneyBadger, Maidsafe and several others.

Where do you see CasperLabs in 3 years? Where do you see Bitcoin in 3 years? Can you call yourself a bitcoin enthusiast?

Ideally, we see ourselves as a preferred choice for developers who want to create great end-user experiences and do it on a chain that does not sacrifice decentralization and security.

I think bitcoin will be significantly more adopted in 3 years. I think as a store of value it's a great (if inefficient) protocol. I am a bitcoin enthusiast for a long time and started buying it in 2012

What are your relations with governmental structures? Are you pro-regulation or not?

We, of course, comply with all the regulations within our domicile (Switzerland). I  am neither "pro" nor "anti" regulation. I mean on one extreme I think we are all happy that murder is illegal. That said I strongly believe that access to the financial system and global compute resources need to be less restrained and that the public should be able to own, govern and run these entities in a decentralized manner. This is an alternative to the current system, not a replacement.

Do you have offices outside of the US?

We are headquartered in Zug Switzerland. Most of our staff are based in Europe. But we have people from 11 countries that speak a collective 18 languages

Do you plan to do more equity rounds for CasperLabs soon?

We are always open to various instruments, currently, our focus is on bringing on validators through the validator sale

Did any major players show interest in the CasperLabs?

Yes, we had a $14.5M Series A which had extremely well-regarded funds including Arrington XRP, Acuitas, Hashkey, Blockchange (Ken Seiff, Gavin Wood), Hyundai etc.

We have several major staking pools that have participated in our testnet and validator sale and are closing out a few more. Stay posted for more announcements on the staking side. Feel free to join our telegram to keep track of all announcements.

Thanks, everyone once again for participating in the AMA. Thanks, Mrinal for sharing your knowledge, it was a pleasure. 

Don’t forget to visit their website at https://casperlabs.io.