AVDHESH CHARJAN · SGP · TRADER-DEV

Now

NTU Computer Engineering, final year (minor: Economics). Open for the next role.

Research

Polymarket market-making economics. EVM client internals. MCP tooling. Crypto × AI.

Stack

TypeScript · Python · Solidity · Rust · C/C++ · Java · React · Next · Node

Chains

Ethereum · Solana · Kaia · Polygon (Miden, Plonky3)

Last

Research Analyst @chainofthought (Aug–Oct 2025) · SWE @monstre · SWE @kaiachain

IRL

Muay Thai (NTU fight team) · biryani on Twitter, boiled potato in person

I've been thinking about trust for a long time. Longer than is probably healthy. The conclusion I keep landing on is that I trust math more than I trust people. Not because people are bad. Because there are too many of them, and at scale, character is a coin flip.

Trust is what holds everything together. Relationships, civilisations, markets. Love is what we call trust after it's already in place. The hard part is getting there, and most of the systems we built to manage it are run by people who fail at scale. 2008 was one version of that. Every captured platform since has been another. "Just trust us" stopped being a strategy a while ago.

What I find interesting about trustless systems is that they don't ask you to be virtuous. Bitcoin, zero-knowledge proofs, P2P networks. They make verification cheaper than betrayal. You can prove something is true without having to vouch for the person saying it. That's a real shift, and I think it's the only honest place to build from.

Free networks and free markets fall out of the same idea. Permissionless. Competitive. Survivable when the people inside are selfish or lazy or actively trying to break it. Markets, as far as I can tell, are discovery mechanisms. They aren't moral systems and they don't pretend to be. They work when entry is open, exit is real, and risk doesn't mean ruin. Anything that needs permission to join is a control surface. Anything that restricts who gets in is a rent machine.

If a network can't survive its adversaries, it doesn't deserve users. If freedom is optional, it's already gone.

That's the why behind what I work on. Microstructure on Polymarket, because it's one of the most adversarial venues I can find with public data. EVM client internals, because someone has to actually read the code if any of this is going to keep working. Agent tooling, because the next decade of cooperation is going to run on protocols, not promises.

trust > love (jul 2024)  ·  free networks and free markets (dec 2025)

  COMMITS  (LIFETIME)              530
  COMMITS  (YTD 2026)              158    ▲ 74% of 2025 in 4 months
  COMMITS  (2025)                  213
  REPOS    (ORIG)                   60
  REPOS    (YTD 2026)                6
  CLIENTS  EXPLORED (FORKS)         40    reth · lodestar · nethermind · ethereumjs · +36
  ERA                          YEAR 7    since 2019-04-04