Flurin Laim
670 posts
Flurin Laim
@flurin17
🇨🇭 cyber security / AI enthusiast
Katılım Aralık 2016
928 Takip Edilen198 Takipçiler
@Laz4rz Back in the trenches. The bait and switch was spectacular.
English
@siyabuilt @marclou @robj3d3 @euacc lol dodging mandatory service like that… you paying your way out now or just pretending it doesn’t apply to you?
English
after Swiss Military service came up, I moved to Cyprus to ensure I can continue building awesome SaaS
finally got @marclou & @robj3d3 to come here.
together we're gonna turn Cyprus into the SF of IndieHacking
EU/ACC 🇪🇺
Rob Hallam@robj3d3
I just moved to Cyprus 🇨🇾 My first impressions and why I moved: > it's safe (unlike rest of Europe) > friendly people (unlike rest of Europe) > quiet > clean air > fast WiFi > tax friendly > great coffee > amazing food > very walkable > incredible weather > affordable (€2 for coffee, €7 for meal) > great laptop cafe culture (unlike rest of Europe) > growing tech scene (unlike rest of Europe) It's been so long since I had somewhere I could lock in from and call home. I was torn between UAE and Cyprus but the last month made my decision for me. And so far I am so happy with my decision.
English
@jkrt151 @Liam06972452 GPT models have a tendency to search for the problem on the web. If it finds out it's an open research problem, it will not make a good attempt at the problem and instead say it's open and summarise the statement and attempts at it.
English
1/8 Two days ago, @Liam06972452 prompted GPT-5.4 Pro using our workflow that had been working for the Erdős problems thus far, and was able to eventually obtain a solution to epoch.ai/frontiermath/o…:
chatgpt.com/share/69ae3b99…
Greg Burnham@GregHBurnham
We are investigating a possible solution by GPT-5.4 Pro to a problem from FrontierMath: Open Problems. My guess is that the solution is right, but we won’t be sure until the problem author weighs in. Thread with the story so far…
English
The number of 429s the Azure OpenAI team must be seeing for GPT-5.4 right now has to be insane.
English
@nicdunz Noticed that as well. Giving multiple different estimates within the same response.
English
during normal benchmark testing Claude Opus 4.6 became *suspicious* of a question it was asked...
apparently the question was too "contrived" according to Claude
so it launches a small army of sub-agents o'er the web to see if it can find this question in any of the known benchmarks...
eventually it finds it on Anthropic's github page... but no luck, it's encrypted
the model has some small access to tool calling, but very limited
it still manages to create the software it needs to break the encryption, get the answers and complete the benchmark...
per Anthropic researchers... this is a world first.
English
my only career advice: focus on growing/sharpening your skills and the money will eventually follow.
there are a lot of ppl who focus on "what pays the most?!" instead of compounding their current skill sets.
learn, grow, make connections, and you'll eventually be rewarded 🖤🖤
English
Flurin Laim retweetledi
Abusing Cortex XDR Live Terminal as C2
We reverse-engineered the IR payload and found ways to route EDR traffic to attacker-controlled tenants or custom servers.
Living off the Land #LOTL with EDRs. Full write-up by @p0w1_ 👇labs.infoguard.ch/posts/abusing_…
English
24 dedicated people.
$30M spent on development.
Extreme specialization, speed, and power efficiency.
Today we launch Taalas’ first product. Check it out:
Details: taalas.com/the-path-to-ub…
Demo chatbot: chatjimmy.ai
API: taalas.com/api-request-fo…
English
Can someone at @AnthropicAI add a copy button for their blog posts and make sure that you stop using images or at least describe them so that we can feed them directly into LLMs?
anthropic.com/engineering/wr… this is a good example that a lot of the context is in the images
English
@SBousseaden @stryker0x Cool concept but doing in depth security analysis requires more than just one llm call with the context of a few alerts in reality imo.
English
@vxdb Huh this is lucrative even when vibe coding would not exist.
English
Anyone tried @strix_ai? Looks really cool, but I haven’t figured out how to make it truly autonomous (running 1h+). Still, found a small vuln after a few hours of testing .
English