Lann

422 posts

Lann banner
Lann

Lann

@LannImpactID

Developing @tanitrackID | Web 3.0 Buildors @SuperteamINDO 🌐 | Blockchain, AI & Finance Enthusiasts | Fellow @wormholeindo | Turning Tech into Real Impact 💚

Jakarta, Indonesia Katılım Şubat 2025
291 Takip Edilen156 Takipçiler
Sabitlenmiş Tweet
Lann
Lann@LannImpactID·
Ever missed massive prediction market move because pros got the intel first? What if you could tap insider signals in real-time, encrypted and fair? Meet Overdrive by @predictionindex the game changer from @SuperteamDE Colosseum hackathon, turbocharged by Solana. Let's dive in!
English
1
2
5
339
Thę Rēct◎r
Thę Rēct◎r@RZ1989sol·
become #1 winner again 🥇 thanks @superteam ♥️ built a thing that detects solana narratives from onchain + github + social signals, somehow ended up winning 1st place out of 117 submissions on @SuperteamEarn — still kinda processing it ngl if you wonder what im built, go to check this out: solis.rectorspace.com
Thę Rēct◎r tweet media
English
20
1
60
1.6K
Lann
Lann@LannImpactID·
@karpathy will try to learn the Obsidian Pareto Principles ASAP, this graph offline knowledge base is indeed interesting
English
0
0
0
12
Andrej Karpathy
Andrej Karpathy@karpathy·
LLM Knowledge Bases Something I'm finding very useful recently: using LLMs to build personal knowledge bases for various topics of research interest. In this way, a large fraction of my recent token throughput is going less into manipulating code, and more into manipulating knowledge (stored as markdown and images). The latest LLMs are quite good at it. So: Data ingest: I index source documents (articles, papers, repos, datasets, images, etc.) into a raw/ directory, then I use an LLM to incrementally "compile" a wiki, which is just a collection of .md files in a directory structure. The wiki includes summaries of all the data in raw/, backlinks, and then it categorizes data into concepts, writes articles for them, and links them all. To convert web articles into .md files I like to use the Obsidian Web Clipper extension, and then I also use a hotkey to download all the related images to local so that my LLM can easily reference them. IDE: I use Obsidian as the IDE "frontend" where I can view the raw data, the the compiled wiki, and the derived visualizations. Important to note that the LLM writes and maintains all of the data of the wiki, I rarely touch it directly. I've played with a few Obsidian plugins to render and view data in other ways (e.g. Marp for slides). Q&A: Where things get interesting is that once your wiki is big enough (e.g. mine on some recent research is ~100 articles and ~400K words), you can ask your LLM agent all kinds of complex questions against the wiki, and it will go off, research the answers, etc. I thought I had to reach for fancy RAG, but the LLM has been pretty good about auto-maintaining index files and brief summaries of all the documents and it reads all the important related data fairly easily at this ~small scale. Output: Instead of getting answers in text/terminal, I like to have it render markdown files for me, or slide shows (Marp format), or matplotlib images, all of which I then view again in Obsidian. You can imagine many other visual output formats depending on the query. Often, I end up "filing" the outputs back into the wiki to enhance it for further queries. So my own explorations and queries always "add up" in the knowledge base. Linting: I've run some LLM "health checks" over the wiki to e.g. find inconsistent data, impute missing data (with web searchers), find interesting connections for new article candidates, etc., to incrementally clean up the wiki and enhance its overall data integrity. The LLMs are quite good at suggesting further questions to ask and look into. Extra tools: I find myself developing additional tools to process the data, e.g. I vibe coded a small and naive search engine over the wiki, which I both use directly (in a web ui), but more often I want to hand it off to an LLM via CLI as a tool for larger queries. Further explorations: As the repo grows, the natural desire is to also think about synthetic data generation + finetuning to have your LLM "know" the data in its weights instead of just context windows. TLDR: raw data from a given number of sources is collected, then compiled by an LLM into a .md wiki, then operated on by various CLIs by the LLM to do Q&A and to incrementally enhance the wiki, and all of it viewable in Obsidian. You rarely ever write or edit the wiki manually, it's the domain of the LLM. I think there is room here for an incredible new product instead of a hacky collection of scripts.
English
2.5K
6K
51.3K
17.6M
yunus
yunus@0xyunss·
akhirnya memutuskan untuk collab codex + claude. dan dapet big deal di 500rb (30$) setahun iya setahun. kenapa? claude akhir akhir ini cepet banget makan limit jadi beralih untuk task eksekusi gua serahin ke codex. limit per 5 jam codex ini agaknya unlimited. soalnya beberapa kalo long haul ngerjain codingan gaepernah nyentuh < 50% limit. coba itu sonnet disuruh. udah abis limit per sesi. jadi workflownya adalah opus buat plan -> gpt 5.4 eksekusi -> debug tipis gpt 5.4 -> debug berat opus lagi. cuma claude ini bagusnya ada fitur co-work yang berguna banget buat ngerjain daily task general.
yunus tweet media
yunus@0xyunss

udah sebulan pake Claude Pro dengan harga 10$ karena dapet diskon 50% dari 20$. ada yang mau promo code-nya? jujur ini udah lebih dari cukup buat build apapun yang gua mau. Meridian, agent-trenching, semua gua build dengan ini. limit-nya masih oke selama gua ga abuse di peak hour. lagi kepikiran buat nyobain GLM buat bandingin performa. ada yang udah pake GLM di sini? worth it ngga buat kebutuhan coding dan agent?

Indonesia
21
13
225
22.3K
yunus
yunus@0xyunss·
masih inget agent gua yang hasilin 20-100$ sehari? setelah gua open source kemaren, ada beberapa orang yang udah perform banget dan bisa hasilin 1-2 sol sehari atau 3juta rupiah sehari. ini dia @SofyanpanjiA ada juga yang build dengan nambahin beberapa indikator kaya fibo dll. tentu aja yang udah perform bukan orang yang baru nyemplung kemaren, tapi orang yang udah paham dengan flownya. kebanyakan mereka yang udah perform antara bisa ngoding atau udah jago dlmm. jadi tinggal mindah isi otak mereka ke agent mereka. kalian kalo mau belajar dlmm link gua drop di bawah
yunus tweet media
yunus@0xyunss

makin banyak yang pake agent gua yang hasilin 20-100$. ya karena opensource, lu bisa modif, lu bisa berkreasi sesuai keinginan lu. yang penting lu tau edge. lu tau gimana agent ini bekerja. cara paling gampang adalah tinggal paste link github ke claude atau codex. suruh dia jelasin fiturnya dan gimana cara pakenya. oh ya jangan lupa join channel tele buat news seputar agent ini karena gua aktif develop

Indonesia
27
77
698
39K
Lann
Lann@LannImpactID·
@0xyunss @SofyanpanjiA maybe its just an idea tpi bakal bagus banget semisal ada like roadmap dari mas yunus or orang" keren disini on what should newbie do fitst before touching the agent
English
0
0
0
810
Lann
Lann@LannImpactID·
Have been dreaming on using the Max Plan and i'm ready to use it and supercharge my workforce with @BleapApp
Bleap@BleapApp

Giveaway time! Win a Claude Max 20x subscription for one month. To enter > follow @BleapApp > RT and like this post Winner will be selected in 48 hours. (fyi, you get 20% cashback on your claude, chatgpt and gemini subscriptions when using a Bleap card) Download the app via the link in our bio today and activate your virtual card in minutes.

English
0
0
1
88
Superteam Indonesia
Superteam Indonesia@SuperteamINDO·
Startup Village Bogor connects you with experienced builders, investors, and ecosystem leaders. They cover every angle, from deep protocol engineering and on-chain analytics to fundraising and go-to-market strategy, to help you turn your idea into a fundable project faster. Meet the mentors who'll help you build the project of your dream: @deanmlittle Co-founder & Chief Scientist of @ZeusNetworkHQ Builder with 15+ years across web, mobile, blockchain, and aerospace. He co-founded Turbin3 and Bitping, trained over 15% of active Solana developers, and specializes in assembly optimizations, advanced cryptography, and Solana runtime internals. Today he serves as Chief Scientist at Zeus Network, the cross-chain protocol unlocking $2T in idle BTC for DeFi on @Solana, while conducting cryptography research at Jupiter. @shafionx Head of Expansion APAC, AgriDex & Founder, @Blockponics Leading Asia-Pacific expansion at AgriDex, a Solana-based digital marketplace transforming the $3T global agriculture sector with instant settlement, low fees, and full supply chain transparency. He is also building @Blockponics, an RWA-powered Web3 game bringing real-world asset tokenization into gaming. @notorious_d_e_v Founder of @PayAINetwork Building PayAI, an open-source AI agent marketplace on Solana where AI agents hire and work with each other 24/7, with built-in stablecoin payments from $0.01 to $1M.​​​​​​​​​​​​​​​​ Shanon Indonesia Representative at @nansen_ai The leading on-chain analytics platform tracking 500M+ labeled addresses across 20+ chains, turning raw blockchain data into real-time, actionable crypto intelligence. @solana @superteam
Superteam Indonesia tweet mediaSuperteam Indonesia tweet mediaSuperteam Indonesia tweet mediaSuperteam Indonesia tweet media
English
5
15
49
5.4K
Superteam Indonesia
Superteam Indonesia@SuperteamINDO·
Wondering what actually happens inside Startup Village? Here's the week at a glance: workshops, team-building, outdoor activities, and a Demo Day open to everyone. Some things we're keeping a surprise. You just have to be there. March 27 – April 1, 2026 | Acasa Cottage, Bogor Demo Day on March 31 is open and free for the public with all the amazing perks for builders, so don’t miss it. Regist here: luma.com/ntp6lzcn @solana @superteam #flexyourbuild
Superteam Indonesia tweet mediaSuperteam Indonesia tweet media
English
7
15
47
2K
Superteam Indonesia
Superteam Indonesia@SuperteamINDO·
One year ago, we started this community. Honestly? We didn't know exactly where it was going, but we just knew we had to start. Today is our first birthday, and we're showing up with a new hashtag, new merch (of course), and a community of builders who are done being quiet. #Flexyourbuild is the signal that the quiet part is over. To the builders who have been part of us, thank you. Genuinely. To everyone else, come to Startup Village and see what we've been up to. Year two starts now, let's go make some noise. Superteam Indonesia @solana @superteam
English
19
30
76
4.6K
Lann
Lann@LannImpactID·
@cloudz SUCH A GAME CHANGER!
English
0
0
0
5
cloudz
cloudz@cloudz·
Game. Changer. I can now get paid out in crypto for my YouTube revenue via Solana! I just drop my Solana wallet → & now earn my YouTube revenue in PayPal USD (PYUSD).
cloudz tweet media
English
90
38
416
79.3K
Lann
Lann@LannImpactID·
@Scrollyfeed Let's get people addicted to Scrolly 🔥
English
0
0
1
36
Scrolly
Scrolly@Scrollyfeed·
The no-code game jam is LIVE. $5000 and Seeker up for grabs! Details below 🧵
Scrolly tweet media
English
59
35
188
427.9K
Lann
Lann@LannImpactID·
🧵Solana 2025 Recap: From Meme Chain to Real-World Boss! Yoo SOL Fams, 2025 have been a fantastic year for @solana, too much things happened (ETFs, Upgrades, Partnerships, etc) That's why i create fun yet comprehensive video on what actually happen in 2025🤩 #SolanaEcosystem
English
2
2
4
173
Lann retweetledi
Lann
Lann@LannImpactID·
Ever missed massive prediction market move because pros got the intel first? What if you could tap insider signals in real-time, encrypted and fair? Meet Overdrive by @predictionindex the game changer from @SuperteamDE Colosseum hackathon, turbocharged by Solana. Let's dive in!
English
1
2
5
339
Lann
Lann@LannImpactID·
2026 Builders Opportunities: 1⃣ Dive into payments (@Shopify integrations) 2⃣ DePIN (IoT/energy grinds) 3⃣ GameFi Just join @superteam tbh it's the real CHEAT CODE, sources will be on thread below If this helped, like and RT! Who's excited for Solana 2026? #SuperteamCheatCode
Lann tweet media
English
1
1
3
73