Aashish Methani

225 posts

Aashish Methani

Aashish Methani

@AashishMethani

Dreaming @goRevoke and Designing @SentientAGI

Katılım Kasım 2018
401 Takip Edilen79 Takipçiler
Sabitlenmiş Tweet
Aashish Methani
Aashish Methani@AashishMethani·
we have built sth we believe every onchain user should be using to protect against approval exploits. For almost every defi activity, approval permssion is given that can be exploited if not revoked in time. do join the waitlist and share it with your onchain frens!
gorevoke@gorevoke

Do you love buying memecoins? swapping tokens? minting NFTs? just degen-ing around? chances are - you’ve already given permission to be drained. Introducing GoRevoke - a smarter way to protect your tokens against approval exploits. ✅ no keys shared ✅ no more "I'll revoke it later" ✅ no more "I forgot to revoke" just set it up and relax.

English
1
2
6
545
Aashish Methani
Aashish Methani@AashishMethani·
Complaint reference: RNP260429074734. I have repeatedly asked for a proper inspection, a clear basis for the charges, and a fair resolution, but have still not received one. Resolve this ASAP. I will not be more patient on this.
English
0
0
2
20
Aashish Methani
Aashish Methani@AashishMethani·
@LGIndia @LGE_Global Unacceptable service. I have been following up for over a month and still have no resolution. I already paid for the AC installation, and now I am being asked to pay again for an issue arising from your technician’s work. Neither the Service Manager nor the Service Center has provided a fair resolution. #ConsumerRights #CustomerService
English
4
3
4
53
Aashish Methani
Aashish Methani@AashishMethani·
Taste is not a .md file, its a MCP server
English
0
0
0
15
Aashish Methani
Aashish Methani@AashishMethani·
@Tangem × tennis Three courts. Three legends. Three memorable shots. The idea was simple, what if I could hold a championship point in my hands while paying? Each card captures the most iconic winning shot from the greatest player on each surface.
Aashish Methani tweet media
Tangem@Tangem

It's World Art Day. And we want YOU to design the next Tangem card. Top 5 designs win from a pool of $500 in BTC, and the winning design stands a chance to be printed as a real physical Tangem card. How to enter: 1. Design your card 2. Quote tweet this post with your design and #TangemArtDay 3. Make sure to tag us @Tangem Deadline: April 22, 10pm CET. The canvas is yours.

English
1
0
4
3.2K
Aashish Methani
Aashish Methani@AashishMethani·
Claude design is so good it’s a bit terrifying. Makes me want to design things so distinctly human that an AI would never think. Exciting times 💪
English
0
0
0
71
Aashish Methani
Aashish Methani@AashishMethani·
Existing tool to manage token approvals is reactive. They will always lose to an exploit that moves faster. With GoRevoke we have stepped on realms of proactive onchain security.
gorevoke@gorevoke

GoRevoke Interactive Sandbox is now live 🚀
 We make managing token approvals safer and effortless. Automate your token revocations with confidence.

No more stress about exploits. Set your permissions once, and let GoRevoke keep your approvals in check.

English
0
0
2
100
Aashish Methani retweetledi
Tu Vu
Tu Vu@tuvllms·
🚨 New paper 🚨 Excited to share our new work on EvoSkill (led by @salahalzubi401), a self-evolving framework that automatically discovers and refines agent skills through iterative failure analysis 🔁🧬. It achieves state-of-the-art performance on @databricks's OfficeQA. Check it out! 📰: arxiv.org/abs/2603.02766
Sentient@SentientAGI

Introducing EvoSkill: a framework that analyzes agent failures and automatically builds the missing skills, leading to rapid improvement on difficult benchmarks and generalizable skills across use-cases. +12.1% on SealQA +7.3% on OfficeQA (SOTA) +5.3% on BrowseComp via zero-shot transfer from SealQA Read more below 🧵

English
6
37
123
17.2K
Aashish Methani retweetledi
elvis
elvis@omarsar0·
A self-evolving framework to discover and refine agent skills. Most agent skills I see today are hand-crafted or poorly designed by an agent. Multi-agent systems for building skills look promising. This paper introduces EvoSkill, a self-evolving framework that automatically discovers and refines agent skills through iterative failure analysis. EvoSkill analyzes execution failures, proposes new skills or edits to existing ones, and materializes them into structured, reusable skill folders. Three collaborating agents drive the entire process. An Executor that runs tasks, a Proposer that diagnoses failures, and a Skill-Builder that creates concrete skill folders. A Pareto frontier governs selection, retaining only skills that improve held-out validation performance while keeping the underlying model frozen. On OfficeQA, EvoSkill improves Claude Code with Opus 4.5 from 60.6% to 67.9% exact-match accuracy. On SealQA, it yields a 12.1% gain. Skills evolved on SealQA transfer zero-shot to BrowseComp, improving accuracy by 5.3% without modification. I will continue to track this line of research closely. I think it's really important. Paper: arxiv.org/abs/2603.02766 Learn to build effective AI agents in our academy: academy.dair.ai
elvis tweet media
English
26
71
414
76K
Aashish Methani retweetledi
Sentient
Sentient@SentientAGI·
We’re excited to announce our official partnership with @askalphaxiv to advance open-source research together. alphaXiv is one of the leading research platforms making it easier than ever to discover and parse through cutting-edge work in an era where AI research is progressing at a blistering pace, ushering in a new age of open academic discovery. Together, we’ll host ongoing community talks and spaces around open-source research, model security, and agent architecture, helping builders learn directly from the researchers and engineers pushing the frontier forward. Also, the alphaXiv community will be directly involved in bringing their research ideas to life through Sentient’s open-source foundation. As alphaXiv evolves into a true hub for global research, where you don’t just read papers, but also leverage datasets, benchmarks, and run easily reproducible experiments—we’re excited to partner with them and accelerate open-source AI development. Our shared goal is simple: make open research the fastest path to real progress. Open-source AI 🤝 open research.
English
972
374
2.1K
450.2K
Ayush
Ayush@ayushbherwani·
@AashishMethani that’s why when we hack together, i believe your gut feeling 😂
English
2
0
2
116
Aashish Methani retweetledi
Sentient
Sentient@SentientAGI·
Sentient’s mission is to ensure that Artificial General Intelligence is open-source and not controlled by any single entity. The GRID is on its way…
English
1.6K
31.1K
35.5K
1.2M
Aashish Methani retweetledi
gorevoke
gorevoke@gorevoke·
over $250m has been lost to infinite or forgotten token approvals in the last 3 years. ➡️ 2024: $80m ➡️ 2023: $142m ➡️ 2022: $31m all because you forgot to click a button every once in a while. you sure your approvals are clean?
English
2
8
16
868

Keşfet

@LGIndia @LGE_Global @Tangem @salahalzubi401 @databricks @askalphaxiv @ayushbherwani @ishahbaz