archis.eth 🦇🔊

I’ve been diving deep into memory engines for LLMs — specifically how far we can reduce token usage while improving long-context accuracy. So I built a memory layer for LLMs and ran it against the full LongMemEval-S benchmark: Thanks to @supermemory and @mastra for their benchmarking framework which i used a reference to set up mine 500 questions gpt-4o actor + gpt-4o judge Best-of-3 judging passes Results: 82.22% unweighted accuracy 80.80% micro accuracy Per-category breakdown: • SSU: 92.86% • SSA: 94.64% • SSP: 70.00% • KU: 88.46% • TR: 76.69% • MS: 70.68% The system was strongest on single-session factual recall, and weakest on multi-session aggregation + temporal reasoning

While everyone’s chasing for virality - you chase for performance Analyse and find videos that would actually help you onboard customers Try out today at : ugcs.farm/mcp

@pranshurastogii @PushChain @harshrajat @richajoshi @sumithprabhu_ @m_AmanGupta @0xNilesh @zaryab_eth @0xSiddesh @pritisinghhhh @arunavaBarua_31 all the best for your future endeavours ser 🫡 a heartfelt thanks to you for giving me my first ever breakthrough in web3 through @PushChain 🫶

Some news: after an incredible run, I'm closing my chapter at @PushChain This one's personal.

I’ve been diving deep into memory engines for LLMs — specifically how far we can reduce token usage while improving long-context accuracy. So I built a memory layer for LLMs and ran it against the full LongMemEval-S benchmark: Thanks to @supermemory and @mastra for their benchmarking framework which i used a reference to set up mine 500 questions gpt-4o actor + gpt-4o judge Best-of-3 judging passes Results: 82.22% unweighted accuracy 80.80% micro accuracy Per-category breakdown: • SSU: 92.86% • SSA: 94.64% • SSP: 70.00% • KU: 88.46% • TR: 76.69% • MS: 70.68% The system was strongest on single-session factual recall, and weakest on multi-session aggregation + temporal reasoning

The architecture ended up being a mix of retrieval, reranking, and fallback reasoning: Retrieval side: • Hybrid dense + BM25 fusion • Heuristic + cached LLM query expansion • MMR diversity • Recency bias for knowledge-update questions • Time-anchor bias for temporal questions • Optional cross-encoder + LLM listwise rerank Answer side: • Chunk-level hits rolled up into session-level evidence • Retrieval-confidence scoring • Low-confidence fallback that injects raw transcript from top sessions • Pluggable single-shot or self-critique synthesizer

The more interesting part was token efficiency. Compared to the full-history Chain-of-Note baseline, on the same 500-question dataset and prompt template: • ~3,200 input tokens/question vs ~14,600 • 78% fewer input tokens • +22pp higher accuracy

Hyperliquid Markets Weekly Heatmap DeFi: +11.4% ✅ Stocks: +11.2% ✅ Layer 1: +4.0% ✅ HIP-3: +3.8% ✅ Indices: +3.3% ✅ Meanwhile, AI: -17.5%. RWA: -5.3%. Commodities: -4.5%. DeFi and Stocks running together while AI gets wiped The market isn't rotating out of crypto. It's rotating out of the narrative - How are you positioning yourself? Source: hyperliquid.goldrush.dev

@0xdhruva @conduitxyz congrats ser!

personal update: I've joined @conduitxyz a couple months back as a rollup engineer, working on rollup infrastructure and scaling Ethereum. I have been enjoying the level of ownership and the caliber of the team here. Excited about what we are building. If you are working in rollups, AA, or interop, always happy to chat!

PRIVATE BETA LIVE Catch what others miss, before it’s too late Trench is here. Lock in, dominate the market 🦾😼

@Sampriti026 nashe kam kro bhai

Drunk tweet - spent so much time finding people with whom I could share belongingness only to realise you born alone you die alone and thus anyone you meet is a treasure in itself.

CompliantLLM (@FiddleCubeAI) detects data leaks into any third-party GenAI tools used across your company. It identifies GenAI-specific attacks to catch breaches in both approved and unapproved AI workflows. ycombinator.com/launches/OBo-c… Congrats on the launch, @kaushik_himself & @nupoor_neha

@daksh0x pop out the key with a guitar pick, and clean it, pretty easy to do i would say, just be gentle while putting it back, there’s some sliding mechanism

folks anyone knows where can I get my macbook's keyboard cleaned? there's something stuck under the 'a' key and i can't seem to get out and it's irritating me. someplace which doesn't burn a hole for something as simple as a keyboard clean and is reliable is appreciated.

Trenchers, stand by! A silent vessel cuts through the dark, armed with what comes next [PRIVATE BETA] is loading…

@nitanshu Should checkout @OfflineProtocol

Noice stuff! Built out couple of projects (with clever integration of low bandwidth encryption + SMS service) during my sophomore days which focused towards offline evaluation of students in remote areas. Solving the commuting issues that students faced during the COVID-19 lockdown.

@daksh0x @gakonst Yo ser, would like to collab and get it to prod

hey @gakonst I was able to vibe code and build a wip version of it. most of the components are being built using crypto primitives. still a wip, not ready for prod. github.com/dawksh/swiem

@prrthamm @thenetworkstate Crazyyy stuff ser

I’m starting a daily video series to share my experience at @thenetworkstate Day 1 at Network School was intense—looking forward to what the next days will unfold

I can literally buy TSLA stock using TRUMP memecoin XD That too even privately Using @encifherio Thanks to @JupiterExchange

You can just buy US stocks privately

@MurrLincoln @coinbase @CoinbaseDev Yeah ive roadmap for scaling the proj to the mainnet, as well as some ideas to improve the latency and throughput of the x402 protocol itself, lmk if i can drop you a DM

@Arch_0125 @coinbase @CoinbaseDev thisi s a great idea, thank you for sharing! let me know if there's anything I can do to support

Played around with @coinbase x402 protocol and here’s what I built recently Codey ✨ - unlock true pay-as-you-go in your AI powered code editors. Use your wallet as API key, top-up USDC & pay-per-call. Pick your context, let smart agents pick the best model for price & accuracy. You control the budget, no flat subscriptions, no wastage. Where to next - Currently wallet is stored in VSCode SecretStorage, can be extended to use CDP wallet with session keys and permissions - Democratising model choice, anyone can host their service and provide api endpoints and set their fees Check out the demo, DM to get access and try out cc @CoinbaseDev