Sabitlenmiş Tweet
raunit
6.3K posts
raunit
@raunit
i build on solana prev: @sendaifun
nyc(one day) Katılım Eylül 2024
756 Takip Edilen1.8K Takipçiler
🚨 BIG STATEMENT BY HUMAYUN KABIR :
"What has happened is good"
"TMC surpassed the loot which was done by the British looted in 100-200 years" 😳
"Mamata gave so much authority to her nephew"
"They cheated the people and looted their money"
"I congratulate the people of West Bengal who gave such a reply to TMC"
News Algebra@NewsAlgebraIND
WEST BENGAL CURRENT TRENDS - BJP : 191 TMC : 97 CONG : 1 LEFT : 2
English
@jacobvcreech - Subsidize vote fees temporarily through foundation/client incentives, especially for small validators
English
Make slots times as short as possible asap. Hear me out.
Pros:
- Better for Market Makers
- Best in class UX for any app
- Continue to be better than any chain including centralized ones
- Faster response times
Cons:
- Harder to index
- Temporarily more vote fees until alpenglow
Worth the tradeoff?
English
POD FACT: the Pod is all data. There is no padding. #podfacts
Solana Events@SolanaEvents
We're excited to announce The Pod. 3 hours of technical programming featuring the deepest dives, happening alongside Accelerate. p-token, post-quantum, RPC 2.0, Quasar, agentic infra, security, and more. If you're a dev working on Solana's hardest problems, this is for you. Powered by @Rockaway_X
English
Update: I was able to backtest it it’s possible, but only to a limited extent. I can test a few strategies, but it won’t scale.
I have another idea to build on this, so I’ll try to create a POC over the weekend and will share an update
Let’s see how it goes.
raunit@raunit
building a backtesting infrastructure, starting with Phoenix as a proof of concept. If it works, I’ll scale it to other protocols I was about to post “someone should build this,” then I remembered this tweet: x.com/GuiBibeau/stat… The idea is clear it just needs some iterations.I’ll use Claude, since I’ve hit the Codex limits. For those unfamiliar, here’s what backtesting means: If a trader wants to take a trade at time t+2, I can simulate or evaluate that trade using data from time t. In other words, it’s based on real-time data, but relative to when you’re placing the trade, it’s treated as historical even if it’s just a few slots behind. Simply put: “historical” means anything that has already happened on-chain and is recorded in the network.
English
Today we’re announcing our $5 million seed round, led by @Multicoin
This funding directly supports Exponent’s next phase: becoming the premier platform for onchain portfolio construction
English
I built an AI friend called Locky.
It has one goal, to help you actually build the thing you keep putting off.
You can @textlocky directly on iMessage.
Here's what it can do today:
English
@thsottiaux please ship /remote control I want to check my Codex session from my phone, even while I’m in the shower 🥲
English
raunit retweetledi
When someone sends you USDC, your wallet address never appears in the transaction's account keys. The transfer goes to your ATA.
Stream every transaction through a Yellowstone gRPC interface on your wallet pubkey, and you miss it entirely.
The user typically doesn't know all of the wallet's ATAs upfront, which is the real problem.
gTFA solves it in one call. It resolves which token accounts belong to a wallet, returns transactions whose balances actually changed, and filters out noise where the ATA was touched, but nothing moved.
Comparing it to a streaming pipeline and you're processing every transaction on the chain, resolving ATA ownership, diffing token balances, dropping no-ops, and building your own index.
Alpha: Stay tuned for another huge release from Helius on this front today, can't wait for our competitors to refute it by writing a blog.
nick | helius.dev@nick_pennie
the value prop of getTransactionsForAddress (gTFA) is NOT the historical data – you can get that from RocksDB snapshots already gTFA is an index it's purpose is reduce dev time, reduce on-call work, and get your product to market faster let's say you want wallet token history for a specific month with raw data, you need to: 1. resolve the time range to epochs 2. fetch the 1B+ txs for those epochs 3. deserialize every tx 3. track ATA ownership 4. filter out unmatched txs (99%+) 5. parse & aggregate remaining transactions that ends up being too slow for a single query – so you have to build a data pipeline, storage layer, and API gTFA does this with a single query in under 50ms!
English
