Alex Donn

18.9K posts

Alex Donn banner
Alex Donn

Alex Donn

@Alex_Donn

48 65 6C 6C 6F! https://t.co/GisMyT4QVh #machinelearning #collaborationovercompetition ☁——– ✈ do you vibe with collaboration over competition? If so, let's meet 👇

🛸 Katılım Eylül 2007
4.9K Takip Edilen4.7K Takipçiler
Castle Island Ventures
Castle Island Ventures@CastleIslandVC·
Be aware that members of the Castle Island Ventures team are being impersonated on Telegram. We will never ask you for money or schedule meetings via Calendly on Telegram. Don't click links if you are contacted by an impersonator
English
10
7
52
37.7K
Bloqarl | Zealynx
Bloqarl | Zealynx@TheBlockChainer·
@Alex_Donn See you in Cannes! ETHCC is going to be packed this year. On which day are you arriving?
English
1
0
0
39
Alex Donn
Alex Donn@Alex_Donn·
2026 web3 events I'm headed to... are you coming? • Blockchain Summit USA — Mar 16–17 — Miami, USA — blockchainsummitusa.com • ETHCC — Mar 30–Apr 2 — Cannes, France — ethcc.io • GITEX Asia — Apr 9 — Singapore — gitex.com/asia/
English
2
0
1
101
Alex Donn
Alex Donn@Alex_Donn·
wait… Amazon isn’t testing nor reviewing AI generated code?! 😆😆😆
Anish Moonka@anishmoonka

Amazon had four Sev-1 outages (their highest severity level) in a single week. Internal memos say AI-assisted code changes were a contributing factor. The timeline here is wild. In October 2025, Amazon laid off 14,000 corporate employees. In January 2026, another 16,000. That’s about 30,000 people in five months, roughly 10% of the corporate workforce. CEO Andy Jassy said the cuts were about culture, not AI. During those same months, Amazon set a target: 80% of developers using AI coding tools at least once a week. They tracked adoption closely and blocked rival tools like OpenAI’s Codex. Even so, 30% of developers still hadn’t touched Amazon’s in-house tool Kiro by January. In December 2025, Kiro caused a 13-hour AWS outage. The AI tool had production-level permissions and decided the best fix for a bug was to delete and recreate an entire live environment. A second incident involved Amazon Q Developer, another AI tool. Amazon blamed both on “user error, not AI.” But quietly added mandatory peer review for all production access afterward. Then March 5: Amazon’s retail site went down for about six hours. Over 22,000 users reported checkout failures, missing prices, and app crashes. Amazon called it a “software code deployment” error. Five days later, SVP Dave Treadwell made the normally optional weekly engineering meeting mandatory. His memo acknowledged “GenAI tools supplementing or accelerating production change instructions, leading to unsafe practices.” These problems trace back to Q3 2025. Amazon’s own assessment: their GenAI safeguards “are not yet fully established.” The new rule: junior and mid-level engineers now need senior sign-off on any AI-assisted production changes. Treadwell also announced “controlled friction” for the most critical parts of the retail experience. For context, Google’s 2025 DORA report found 90% of developers use AI for coding but only 24% trust it “a lot.” An Uplevel study of 800 developers found Copilot users introduced 41% more bugs with no improvement in output. Amazon is finding out what those numbers look like at the scale of a $500 Billion revenue company, with 30,000 fewer people on staff to catch the mistakes.

English
0
0
2
259
Sirada
Sirada@sirada_l·
Shifted to a new Telegram account (different from my X handle). If you’re unsure whether a message is from my real account, DM me on X to verify. Stay safe from scams (aka fake Sirada).
English
21
0
55
5.2K
Alex Donn retweetledi
OtterSec
OtterSec@osec_io·
We found the same Fiat-Shamir bug in six independent zkVMs. The result: an attacker can bypass the cryptography entirely and prove mathematically impossible statements (like minting $1M out of thin air). Full breakdown ↓
OtterSec tweet media
English
34
139
863
117K
Alex Donn retweetledi
Jay
Jay@jayendra_jog·
This novel approach to sharding transaction data is only available to chains using MCP for consensus, which is why we teamed up with researchers @LefKok and @alberto_sonnino Read more here: arxiv.org/pdf/2512.17045
English
1
1
13
889
Alex Donn
Alex Donn@Alex_Donn·
@Cryptocito nice update 🔥 any cosmos events coming up for q1?
English
0
0
0
10
Cito
Cito@Cryptocito·
Tough times make strong Coins. My updated Crypto thesis for 2026 covering Cosmos, ATOM, key positions I still hold (and want to build) as well as how my conviction is evolving:
English
5
1
25
6.9K
Claude
Claude@claudeai·
Introducing Cowork: Claude Code for the rest of your work. Cowork lets you complete non-technical tasks much like how developers use Claude Code.
English
2.6K
8.4K
86.9K
49.7M
Alex Donn retweetledi
OtterSec
OtterSec@osec_io·
We have just completed our thorough audit of @jup_lend, and we are happy with @JupiterExchange and @0xFluid’s attention to detail and security!
Jupiter Lend@jup_lend

BREAKING: The fourth independent audit of Jupiter Lend’s smart contracts has been successfully completed by @osec_io. • 4 audits completed • 1 additional audit in progress • Up next: open-sourcing the code followed by a @code4rena review We remain committed to improving with community feedback, engaging the top audit firms in the space, and keeping security as a top priority.

English
3
14
82
27.1K
Alex Donn retweetledi
OtterSec
OtterSec@osec_io·
NEW: OAuth misconfigurations show how common dev settings can lead to account takeovers. Our second deep dive breaks down real cases where overlooking differences between desktop and mobile environments left SDKs, exchanges, and wallets open to exploits. osec.io/blog/2025-10-1…
English
2
18
76
8.1K
Alex Donn
Alex Donn@Alex_Donn·
The End of Fine-Tuning 💀 A new Stanford paper introduces Agentic Context Engineering (ACE), a breakthrough that profoundly improves model performance without training a single weight. ACE bypasses traditional fine-tuning by having the model self-evolve its own prompt. The system iteratively writes, reflects on its failures, and edits its context, turning every success into a rule and every misstep into a strategy—creating a truly self-improving system. The results are remarkable: * Smarter: Outperforms GPT-4 agents by +10.6% (AppWorld) and +8.6% (finance reasoning). * More Efficient: 86.9% lower cost and latency. * Agile: Requires no new labels—just feedback. We're shifting from static, fine-tuned models to dynamic, self-tuned intelligence. This is the start of true AI. Read the full paper here: arxiv.org/abs/2510.04618
English
0
0
5
178