pod network

297 posts

pod network banner
pod network

pod network

@poddotnetwork

One Market to Rule Them All

Katılım Eylül 2024
33 Takip Edilen15.8K Takipçiler
pod network retweetledi
shresth
shresth@shresth3103·
markets today allow informed flow to capture the maximum value from the book - making it toxic. informed flow is good. it drives price discovery. the problem is the cost structure. fifo lets the first participant extract the full signal value with no competition. that cost gets passed on - wider spreads, worse quotes for everyone else. batch auctions change the game: if multiple traders have the same information, they compete on price, not speed. whoever is willing to offer the better price gets matched. the signal gets priced at its true marginal value → tighter spreads, less toxicity. fifo → oligopoly
batch auctions → free market
shresth tweet media
Niko@0xniko0x

Every trade you've ever made carried a hidden tax Every time you trade on a conventional exchange, you pay a spread. You know this. What you probably don't know is why it's so much wider than it needs to be. Traders who act on better information push prices toward fair value. That's not a flaw - that's price discovery working exactly as it should. However, in a FiFo market, the fastest participant prices their information before anyone else can react. No competition, hence no cheaper alternative. The cost becomes the maximum value they can extract based on the available liquidity - and the market has no choice but to pay it. First in, first to extract FiFo doesn't reward better information. It rewards faster infrastructure. The first to arrive captures the full value of a signal - even if ten others held the same view and would have offered it cheaper. The signal gets monetized once, by one participant, at full extraction. That's what makes informed flow toxic. Liquidity providers adapt - wider spreads, pulled quotes, reduced size. Every adjustment makes the market worse for everyone else. The spread you pay on every trade includes the accumulated cost of a market defending itself against its own design. A hidden tax, baked into every transaction. Speed is the wrong race On Pod, all orders within a defined window clear at a single uniform price. First or last - it doesn't matter. Speed is structurally irrelevant. This creates a free market, which prices information at its true marginal value. Informed traders compete on the quality of their signal, not their connection. If someone else holds the same view, they'll offer it cheaper. Spreads tighten. The arms race ends. The hidden tax disappears. What was taken silently is returned transparently Every signal has a marginal value. When the market prices that correctly, a natural remainder emerges - the surplus. The gap between what traders were willing to pay and what they actually traded at. In a FIFO market, surplus is non-existent due to the maximum extracted value. On @poddotnetwork it is returned back to every participant. The clearing price isn't set by one trader. It's the product of every order in that batch - from the largest to the smallest trade. Every participant shaped that price, so every one deserves a share of what it produced, proportional to their contribution.

English
1
2
10
534
pod network retweetledi
Niko
Niko@0xniko0x·
Every trade you've ever made carried a hidden tax Every time you trade on a conventional exchange, you pay a spread. You know this. What you probably don't know is why it's so much wider than it needs to be. Traders who act on better information push prices toward fair value. That's not a flaw - that's price discovery working exactly as it should. However, in a FiFo market, the fastest participant prices their information before anyone else can react. No competition, hence no cheaper alternative. The cost becomes the maximum value they can extract based on the available liquidity - and the market has no choice but to pay it. First in, first to extract FiFo doesn't reward better information. It rewards faster infrastructure. The first to arrive captures the full value of a signal - even if ten others held the same view and would have offered it cheaper. The signal gets monetized once, by one participant, at full extraction. That's what makes informed flow toxic. Liquidity providers adapt - wider spreads, pulled quotes, reduced size. Every adjustment makes the market worse for everyone else. The spread you pay on every trade includes the accumulated cost of a market defending itself against its own design. A hidden tax, baked into every transaction. Speed is the wrong race On Pod, all orders within a defined window clear at a single uniform price. First or last - it doesn't matter. Speed is structurally irrelevant. This creates a free market, which prices information at its true marginal value. Informed traders compete on the quality of their signal, not their connection. If someone else holds the same view, they'll offer it cheaper. Spreads tighten. The arms race ends. The hidden tax disappears. What was taken silently is returned transparently Every signal has a marginal value. When the market prices that correctly, a natural remainder emerges - the surplus. The gap between what traders were willing to pay and what they actually traded at. In a FIFO market, surplus is non-existent due to the maximum extracted value. On @poddotnetwork it is returned back to every participant. The clearing price isn't set by one trader. It's the product of every order in that batch - from the largest to the smallest trade. Every participant shaped that price, so every one deserves a share of what it produced, proportional to their contribution.
English
1
2
7
758
pod network retweetledi
shresth
shresth@shresth3103·
global payments => tempo global markets => pod opinionated, purpose built L1s will be 100x better general purpose L1s. why? tradeoffs are much easier to make when you have a clear use case in mind!
English
0
3
10
709
pod network
pod network@poddotnetwork·
Batch auctions allow for fair pricing of illiquid assets. Prediction markets are a perfect example — exotic, low-liquidity assets that need a better mechanism. Pod runs on batch auctions. That makes our infrastructure a natural fit for prediction markets. 👂 Hear Pod's @shresth3103 explain why batch auctions are the right mechanism for prediction markets on @CryptoCoinShow.
English
9
10
36
1K
pod network retweetledi
Diario฿itcoin
Diario฿itcoin@DiarioBitcoin·
🚨 Pod Network se lanza a eliminar el MEV en los mercados cripto. La oportunidad injusta que tienen los secuenciadores y validadores sobre las órdenes de los usuarios será atacada mediante subastas por lotes. Agrawal, CEO de Pod, señala que la desigualdad persiste incluso en plataformas descentralizadas. Con un enfoque en mercados de acciones y activos físicos, su objetivo es asegurar una ejecución más justa y equitativa. La espera es por un sistema que priorice al usuario final y restituya el valor perdido en transacciones.
Diario฿itcoin tweet media
Español
4
6
12
918
pod network
pod network@poddotnetwork·
The Market Is Rigged - Just Not the Way You Think.
Niko@0xniko0x

The Market Is Rigged - Just Not the Way You Think Most people think markets are unfair because of insiders or manipulation. That’s not the real problem. Every trade you make, someone faster is on the other side. They've already won before you hit confirm. You're just paying for it - on every trade, whether you know it or not. 1. The hidden problem of CLOBs Most exchanges run on a Central Limit Order Book (CLOB): real-time matching, tick-by-tick updates, first-in-first-out. Simple. Efficient. But structurally biased. Orders are processed sequentially, so the fastest participant always gets priority. Traders with superior infrastructure or proprietary data access can react first and pick-off stale quotes before the rest of the market adjusts. LPs know this and to protect themselves, they widen spreads and quote more cautiously, pricing in the constant risk of adverse selection. As a result, the end user gets a worse price. The faster you are, the more you extract. Everyone else pays for it. 2. Frequent Batch Auctions on pod Frequent Batch Auctions (FBAs) eliminate this arms race Instead of matching orders continuously, FBA's group orders into very short, discrete time windows (50 -200ms) and clear them at a single uniform price. Because everyone in the batch trades on the same price , speed advantages disappear. The batch window is the speed bump itself - except instead artificially slowing down fast participants, it makes speed structurally irrelevant. A retail trader and a co - located HFT participate on equal terms, removing the constant race for microsecond execution. This single structural change has meaningful downstream effects: - Adverse selection disappears - Spreads tighten - Price discovery improves - Manipulation becomes structurally impossible - Batches are transparent and verifiable 3. What this means in practise On @poddotnetwork the structural tax of wider spreads, thinner books and adverse selection is removed at the protocol level. LPs quote tighter because they carry less risk. Traders execute better because the liquidity they access is genuine, stable, and priced fairly. The market functions the way it should. This isn't just a marginal improvement. FBA's represent a fundamental rethinking of what a fair market looks like - shifting the competitive edge away from latency and toward what should always have mattered: pricing accuracy, capital efficiency, and execution quality for everyone.

English
4
2
23
1.2K
pod network retweetledi
Niko
Niko@0xniko0x·
The Market Is Rigged - Just Not the Way You Think Most people think markets are unfair because of insiders or manipulation. That’s not the real problem. Every trade you make, someone faster is on the other side. They've already won before you hit confirm. You're just paying for it - on every trade, whether you know it or not. 1. The hidden problem of CLOBs Most exchanges run on a Central Limit Order Book (CLOB): real-time matching, tick-by-tick updates, first-in-first-out. Simple. Efficient. But structurally biased. Orders are processed sequentially, so the fastest participant always gets priority. Traders with superior infrastructure or proprietary data access can react first and pick-off stale quotes before the rest of the market adjusts. LPs know this and to protect themselves, they widen spreads and quote more cautiously, pricing in the constant risk of adverse selection. As a result, the end user gets a worse price. The faster you are, the more you extract. Everyone else pays for it. 2. Frequent Batch Auctions on pod Frequent Batch Auctions (FBAs) eliminate this arms race Instead of matching orders continuously, FBA's group orders into very short, discrete time windows (50 -200ms) and clear them at a single uniform price. Because everyone in the batch trades on the same price , speed advantages disappear. The batch window is the speed bump itself - except instead artificially slowing down fast participants, it makes speed structurally irrelevant. A retail trader and a co - located HFT participate on equal terms, removing the constant race for microsecond execution. This single structural change has meaningful downstream effects: - Adverse selection disappears - Spreads tighten - Price discovery improves - Manipulation becomes structurally impossible - Batches are transparent and verifiable 3. What this means in practise On @poddotnetwork the structural tax of wider spreads, thinner books and adverse selection is removed at the protocol level. LPs quote tighter because they carry less risk. Traders execute better because the liquidity they access is genuine, stable, and priced fairly. The market functions the way it should. This isn't just a marginal improvement. FBA's represent a fundamental rethinking of what a fair market looks like - shifting the competitive edge away from latency and toward what should always have mattered: pricing accuracy, capital efficiency, and execution quality for everyone.
English
7
6
30
3.9K
pod network retweetledi
Sam Kamani
Sam Kamani@samkamani·
"Block proposers can reorder transactions to create risk-free profit." - @shresth3103 , Co-founder, @poddotnetwork Did you know block builders can influence trades on-chain? In the full episode, Shresth breaks down MEV, slippage, and how block proposers extract hidden profits in crypto markets. open.spotify.com/episode/0j90U0…
English
5
3
8
752
pod network retweetledi
Crypto Coin Show
Crypto Coin Show@CryptoCoinShow·
“Decentralized markets” still have a pecking order. In this episode, Shresth Agrawal (CEO & Co-Founder, @poddotnetwork) breaks down why MEV + ordering quietly decide who gets the best execution. Key takeaways: MEV acts like a hidden tax on every trade Latency and finality are the real fairness bottlenecks “200ms finality” changes what is even possible for real users Link in the replies.
Crypto Coin Show tweet media
English
13
5
28
2.2K
pod network
pod network@poddotnetwork·
Traditional applications use a database for storage. Every time you need something, you query it — and that back-and-forth is expensive. Pod keeps its state in memory instead. But that creates a persistence problem: what happens when a machine crashes? The answer is an append log — a binary file that only adds bytes at the end, reducing I/O operations along the hot path. When a node restarts, it replays the log and recreates state from scratch — no database required. 👂 Hear Pod engineer Giannis Gkoulioumis explain how Pod uses an append log to solve the persistence problem.
English
10
9
38
2.4K
pod network
pod network@poddotnetwork·
Markets today are built so that a small set of actors can extract value from retail flow. That is true in both traditional finance and crypto. We think that is a market design problem. At Pod, we are building infrastructure where this value cannot be extracted by design. Retail flow is valuable because it contributes to price discovery, and that value should go back to the retail user as surplus — not to intermediaries. 🎙️Hear Pod CEO, @shresth3103 on the @samkamani podcast
English
19
14
72
3.5K
pod network
pod network@poddotnetwork·
🎙️ pod on the @samkamani podcast. @shresth3103 talks about why today's markets are fundamentally unfair — across both centralized and decentralized exchanges — and how pod is rethinking market design from the ground up. Full episode👇
pod network tweet media
English
9
17
65
3.9K
pod network retweetledi
Alex Brezas
Alex Brezas@abresas·
We just released a conversation with Giannis going deep on how Pod is built. One thread to add context: Actors communicate via messages, and those messages are batched. The specific tradeoff we make is to always drain every pending message and process them all together, maximizing throughput without sacrificing latency, since we never wait for more to arrive. Batching has compounding benefits: it groups disk writes, reduces cross-thread synchronization, and enables batch signing. Instead of signing each validator attestation individually, we hash them all together and sign once. Other validators then verify that single signature for the whole batch. Each transaction's own signature is still verified individually. Being this fast means we can't afford to hit disk on every operation. On every batch we write a very fast binary append-only log that acts like WAL in traditional databases, and only if it grows too large do we store a snapshot of the state that all these operations lead to. Think append log has a list of individual payments, and the balances of each person is the resulting state that is only commited to disk only if there a lot of transactions already. For more see the video of our conversation with Giannis.
pod network@poddotnetwork

👀Inside the Pod Episode 2: Achieving >300k TPS with Giannis and @abresas

English
1
6
22
1.9K
GUBA
GUBA@gubatron·
"Pod sends votes and other messages over UDP rather than TCP. With UDP" Very much like we do in libtorrent for the Bittorrent protocol, it's not only efficient but you can punch through firewalls this way, for bittorrent we use an UDP protocol called RUDP, as in Reliable UDP with retry logic and more, in the case of Pod Network it's for performance. "Pod can pack data tightly into a single packet sized to fit within the MTU, avoiding fragmentation entirely, and then send them without expecting an acknowledgment." Love it, MTU = Maximum Transmission Unit, about 1500 bytes. "Data is serialized as raw bytes using a custom, minimal encoding—no JSON, no external serialization libraries. The format is designed so that encoding and decoding are trivial operations." Fuck yes! no BS human friendly crap, pure efficiency. Loving it! And it's all rust if I remember correctly.
pod network@poddotnetwork

x.com/i/article/2029…

English
4
3
19
1.5K