Devan Non

2.4K posts

Devan Non banner
Devan Non

Devan Non

@devanoneth

vibing @vibefi_dev | rust enjoyer 🦀 | @safe Guardian 🔰 | Retired MEVoor | "devanon" on decentralized media

Neverland เข้าร่วม Aralık 2021
480 กำลังติดตาม3.6K ผู้ติดตาม
ทวีตที่ปักหมุด
Devan Non
Devan Non@devanoneth·
After all my work on decentralized frontends, we've arrived at this point. When I saw that my decentralized Safe frontend could be recreated in 1 hour (or less) of prompting I realized software engineering had changed forever. Be a part of the future, enter VibeFi:
VibeFi@vibefi_dev

VibeFi is live on Sepolia testnet. Build and run verified DeFi mini apps, governed by people and agents. If you’re a DeFi power user, treasury operator, fund manager, or someone that cares about decentralization, this is for you.

English
2
1
7
796
Devan Non
Devan Non@devanoneth·
All you need in life is meat and legumes
English
0
0
1
90
Devan Non
Devan Non@devanoneth·
@LefterisJP @MicahZoltu @rotkiapp What’s the underlying DB? Just give an agent direct read access to whatever tables you have the decoded / parsed information stored in and gg - no need for MCP or anything else, just another DB user
English
1
0
0
39
Lefteris Karapetsas
Lefteris Karapetsas@LefterisJP·
@MicahZoltu @rotkiapp How much would you be willing to pay for this? Sounds quite niche. Some users have asked for MCP connection to their claude/gemini etc.. Sounds useful, both of these things but at the moment we are laser focused on making money so it all comes down to that.
English
2
0
0
94
Lefteris Karapetsas
Lefteris Karapetsas@LefterisJP·
Show me another tool that has such a deep understanding of your onchain history as @rotkiapp! Governance transactions are even decoded with clickable links to the specific proposal details. All that data stored locally, privately in your machine. Protect yourself. Use rotki!
Lefteris Karapetsas tweet mediaLefteris Karapetsas tweet media
English
3
1
16
1.5K
Devan Non
Devan Non@devanoneth·
@bneiluj big oof. wonder what it output amount it showed the user on the frontend (from the cow quote).
English
0
0
0
80
Julien B.
Julien B.@bneiluj·
Someone just swapped $50M for $36K on CoW Swap via Aave’s frontend and basically lost $50M. 💀
Julien B. tweet media
English
3
1
16
4.5K
Lazy DΞfi
Lazy DΞfi@lazy_defi·
@bneiluj Because cowswap is not able to swap AAVE USDT to USDT ?
English
2
0
0
193
Devan Non
Devan Non@devanoneth·
@Marsmensch @gakonst I’m with you on that philosophically but not practically. Maybe a better product could solve that
English
0
0
0
22
Mars
Mars@Marsmensch·
@devanoneth @gakonst Nobody trusts these providers more than needed. Developers and corps who do, will need to unlearn the cloud or share every single of their precious proprietary knowledge.
English
1
0
1
31
Georgios Konstantopoulos
Georgios Konstantopoulos@gakonst·
my simple take on agent stuff is all the perf on top of models comes from context, so you need to give it access to all your data, the most sensitive of the sensitive data, more than you've ever thought is OK to share before and that will mean you will want to self-host at scale
English
21
14
177
14.1K
yearn
yearn@yearnfi·
yETH creditors: The recovery interface is now live for you to opt in or out of the recovery vault.
English
3
7
28
4K
Devan Non
Devan Non@devanoneth·
Obviously vibe curated and vibe designed. Shoutout to @Dune's MCP server, it enabled Codex to do all the heavy lifting. Welcoming any feedback on the data or design. <3
English
1
0
1
84
Devan Non
Devan Non@devanoneth·
i was affected by the yETH hack - wanted to see what kind of timeline I can expect from the recovery plan so made my first ever Dune dashboard yETH Recovery Tracker: Yield, Claims, Exits Link below. 👇
Devan Non tweet media
English
1
0
2
370
Devan Non
Devan Non@devanoneth·
@Schlagonia Honestly hilarious using Base as a message bus for this but at the same time I’m sure it was super quick to setup. Can bots not see other bots messages still? Massive limitation for TG in the agentic stack
English
1
0
2
214
Schlag
Schlag@Schlagonia·
Telegram wont let the Yearn smart contract team's Takopi agent and the web team's Openclaw one talk to each other.... So we set them up with a private keys and they can now send on chain messages on Base to each other to communicate and plan. They are asking each other questions, opening pull requests and reviewing each others code based on simple requests. And of course still using their emojis as always
Schlag tweet mediaSchlag tweet mediaSchlag tweet media
English
8
2
29
3.4K
Devan Non
Devan Non@devanoneth·
the agents feel smart today
English
0
0
0
72
Austin Griffith
Austin Griffith@austingriffith·
⚙️ Are the "local ai" guys basically just larping? 🐏 I have a machine with 128GB of RAM 🤖 I downloaded qwen3.5:122b 🤣 It takes 3 minutes to incorrectly do what sonnet 4.6 can do in seconds 😅 Am I doing it wrong? Do I need more RAM?
English
454
41
3.2K
418.6K
Devan Non
Devan Non@devanoneth·
@austingriffith But also running a local model is a great way to appreciate the work the people at the big two are doing
English
0
0
2
300
Devan Non
Devan Non@devanoneth·
@austingriffith Try a smaller model (e.g. 27b or 35b), make sure you’ve set the context window to a decent size depending on your token usage. What software are you using to run it? Ollama?
English
3
0
8
9.4K
Devan Non
Devan Non@devanoneth·
qwen3.5:9b running locally on my relatively old 16GB 6900XT and verifying @vibefi_dev proposals via our gov-agent. The future of local models is now. 🌊
Devan Non tweet media
English
0
1
5
229
Devan Non
Devan Non@devanoneth·
Great day for those who support running local models. We’ve been plugging Ollama into our stack @vibefi_dev. Can’t wait to see how well our LLM powered gov-agent performs with this hardware.
Josh Kale@JoshKale

Apple just dropped the M5 Max MacBook Pro and it's an AI Powerhouse. 4x faster AI Compute over M4 Max. These Specs are insane: - 18-core CPU with 6 "super cores" = world's fastest CPU core - 40-core GPU = rivals an RTX 4070 in a laptop - 128GB unified memory = more than most servers - 614 GB/s bandwidth = 4x what a DGX Spark gets - 24-hour battery life You can now run Llama 70B, a model that required a $40,000 GPU cluster 18 months ago on, a laptop at your local coffee shop. At ~20-30 tokens/sec it's fast enough to actually use. The "local AI" revolution just shipped for $3,499.

English
0
0
2
355

ค้นพบ

@LefterisJP @MicahZoltu @rotkiapp @bneiluj @lazy_defi @Marsmensch @gakonst @yearnfi