really good breakdown of building and scaling a prediction market market maker from scratch. most threads talk about the strategy. this one talks about everything that breaks during testing: stale quotes, websocket desyncs, partial fills, reconciliation, adverse selection, ... also a nice reminder that liquidity compounds. the whole strategy depended on being able to hedge against a deeper, more liquid market nowadays anyone can spin up a bot from their laptop and learn the same market structure lessons that used to require a seat at a prop shop

A milestone for U.S. market structure: a true bitcoin perpetual, approved by the @CFTC on a registered exchange. Perps have long been crypto’s center of gravity for liquidity and price discovery and now that activity is onshore in the U.S. for the first time. This is a huge win for U.S. financial innovation: bringing products onshore and regulated by the @CFTC .

Just in: @jump_firedancer 1.0 is now rolling out.

imc prosperity is unironically a great benchmark for agents; interesting to see how much edge each model can achieve from reading the problem, ingesting the data and one-shotting a solution

Huge congratulations to TradeXYZ and S&P for this historic partnership. I'm honored that these teams choose to build on Hyperliquid. Seeing official S&P500 perpetual futures launch exclusively on Hyperliquid is a validation of everyone's past years of hard work: global access to decentralized finance, perpetual futures as 24/7 price discovery, and Hyperliquid upgrading the existing financial stack to house all of finance. The S&P500 is synonymous with "the market," a single number that captures the essence of the largest economy in the world. Looking forward to tracking the world's most important financial gauge 24/7 on the most liquid permissionless markets.

Super happy that my Prop AMM finally made the leaderboard at #37 with 461 edge It's been extremely valuable to me as a student, and I want to give huge thanks to @danrobinson and @bqbrady for setting up this competition Will continue to fine-tune and improve it!

Im super excited to announce that I will officially be joining the protocol team @Polymarket. Prediction markets are the future. I see a world where anyone who participates in public markets, actively or passively, also participates in prediction markets. Polymarket is the team that’s going to execute on that vision. Not only is the team insanely cracked, but they’re also anchored in the fundamentals that matter: trustless, decentralized markets that scale. Huge thank you to Mike, Joelle, Rodrigo, Jon, Jacob, and Ethan for pushing me through this process. I am super grateful for this opportunity and can’t wait to start building. @_loset Does this mean I can get a Polymarket badge now 😆

I built a native, low-latency C CLOB client for @Polymarket After wrapping up my finals, I’ve been spending my winter break—like a lot of others—trading Polymarket’s 15-minute crypto price markets. Naturally, I started writing bots to trade these markets, and one thing I quickly noticed was that while there were solid Python, TypeScript, and Rust clients, there wasn’t a C option. A lot of trading infrastructure in quantitative trading and HFT is still written in C , and I thought this could be a good on-ramp for folks from that world to experiment with Polymarket using tools they’re already comfortable with, while also being a great learning experience for me as a student. When it comes to Ethereum tooling, Rust, Python, and TypeScript all have fairly mature ecosystems—C is the odd one out. Because of that, I had to implement much of the Ethereum-specific logic completely from scratch, which ended up being one of the most rewarding parts of the project: - EIP-712 typed data signing - RLP encoding - ABI encoding - Ethereum JSON-RPC client - On-chain approval helpers - L2 HMAC auth message construction (Using OpenSSL for cryptographic primitives and libsecp256k1 for signing) Along the way, I spent a lot of time learning and applying low-latency optimizations for high-frequency trading—like keeping TCP/TLS hot, SIMD JSON parsing for fast market data handling via simdjson, background heartbeats, and more. I plan on writing more about this in a future post, so stay tuned. The client is still early, but it already has full feature parity with the Rust client for core trading: - L1 / L2 authentication - Order creation, signing, and posting - Market data endpoints - On-chain approvals (USDC CTF) - All order types (GTC, FOK, GTD, FAK) On the roadmap: - Custom memory allocators - Zero-copy string views - Profile-guided optimization - 100% feature parity with the Rust client I’ve learned a lot from this project, and with the new maker rebates that have just rolled out for 15-minute markets, I expect more participants to start providing liquidity. Hopefully this makes it easier to build tighter, lower-latency trading strategies. School starts soon, but I plan on shipping a lot of the roadmap items before then. Contributors welcome. cc: @PolymarketBuild @shayne_coplan github.com/gnarayan1337/cpp-…

welcome to the future

Happy ETH all time high to all those that celebrate!

phenomenal work

out of everything i've shipped @base, this one's my favorite. cant wait to keep making it better!

back to home @base. mission complete. 🛫🟦 @baseapp

looking forward to showing everyone what we've built

has been an insanely fun experience getting to meet everyone irl and building late into the night

been 2 hours deep tweaking trending coin filters… still playing 😅 fun stuff - thanks for helping me @gnarayan ps: what did you buy today??????

Today me and @gnarayan built hashMind at @MARA hack energy/ai hackathon ⚡️ Its a simulation of an entire cities energy grid, people movement, and compute workloads that finalized as a top 5 contender for the 1 bitcoin prize 🏆

this is huge

sick time at @techroastshow