About

Hi, I’m Sergei!

I am a researcher interested in security, privacy, and scalability of blockchains.

My mission is to advance individual financial sovereignty by combining deep research with pragmatic implementation.

My CV (updated 2023).

The multi-faceted design space of blockchain technologies has fascinated me since 2013. I have contributed to peer-reviewed papers on a range of topics that include:

  • vulnerability detection in Ethereum smart contracts;
  • P2P-level transaction clustering in Bitcoin, Monero, and Zcash;
  • denial-of-service and privacy attacks in the Lightning Network.

I did my PhD at the University of Luxembourg (CryptoLUX group), where I defended my thesis (presentation video) in 2020. I later worked as a postdoctoral researcher at Chaincode Labs. I also co-host Basic Block Radio – a deeply technical blockchain podcast in Russian with 170+ episodes so far.

My research journey

Before 2016: Vulnerability detection

In 2013, I got my Masters degree in applied mathematics and systems programming from the Moscow State University (Faculty of Computational Mathematics and Cybernetics).

In 2013–2016, I was a full-time information security analyst at SmartDec. My tasks included aggregating information about best practices in various programming languages and formalizing dangerous coding patters for our vulnerability detection tool.

I fell into the Bitcoin rabbit hole in late 2013. In my spare time throughout 2014–2016, I wrote for a popular Russian-language cryptocurrency website Bitnovosti, and helped film a documentary about cryptocurrency adoption in Europe.

2016–2017: Bugs in Ethereum contracts

In 2016, I started a PhD program at the University of Luxembourg. My first research topic was the security of Solidity smart contracts. The main results of that period were Findel and Smartcheck.

Findel is a functional domain-specific language (DSL) for financial contracts on top of Solidity. The key idea is to think of a contract as a tree-like structure of elementary operations. The leaves correspond to monetary sums, and the nodes reflect the conditions under which the payments are made. The benefit of a functional DSL, compared to a Turing-complete language, is that it’s easier to analyze and write securely.

Smartcheck was among the first papers on automated security analysis for Solidity code. We proposed a comprehensive classification of bugs in Solidity contracts known at the time, including the infamous re-entrancy vulnerability that destroyed The DAO in 2016. We developed a tool that detects said vulnerabilities, and tested it on a large set of real-world contracts.

2018: P2P-level deanonymization in Bitcoin and friends

In 2018, I studied the P2P layer of Bitcoin and privacy-focused cryptocurrencies (Zcash, Dash, and Monero). The research question was: what information can a well-connected adversary extract from the P2P layer? In the resulting paper, we described a method by which an attacker can cluster transaction that had originated from the same node based solely on their P2P propagation patterns. We successfully clustered our own transactions using patched node software running on geographically distributed servers.

2019–2022: Lightning Network’s security and privacy

In 2019, I became interested in scaling blockchains with second-layer protocols and payment channel networks in particular. I decided to focus on the Lightning Network – the major L2 effort in the Bitcoin ecosystem. During this time, I studied two somewhat related issues: probing and jamming.

Balance probing allows for estimating a remote channel balance by sending unsolicited fake payments. This behavior should not be possible but is hard to discourage, as failed payment attempts are free. We introduced a mathematical model to quantify the amount of information an attacker learns, and applied it to the previously unstudied case of parallel channels.

Channel jamming is a denial-of-service attack where an adversary blocks victim’s channels by initiating payments but not finalizing them. Similar to jamming, the absence of fees for failed payments make attack costs trivial. We proposed a new fee scheme that includes upfront unconditional fees, and measured its effectiveness in a simulation.

I also contributed to a chapter on security and privacy for “Mastering the Lightning Network”.

Publications

My publications with citation counts etc are listed on Google Scholar. My talks and conference presentations are on my YouTube channel.

2022

2021

2020

2019

2018

2017

2016

  • С. Тихомиров, Я. Александров, Е. Марченко, Л. Сафин. Поиск закладок в программном обеспечении (S. Tikhomirov, Y. Alexandrov, E. Marchenko, L. Safin. Finding undocumented features in programs). «Защита информации. Инсайд» №3, 2016 (abstract)

Media appearances

English

Russian