k5hp
Catching Rogue Agents in Real Time with Synchronous Control Monitoring
Synchronous Control Monitoring: Preventing Harmful Agent Actions in Real Time
Runtime Alignment with Synchronous Control Monitoring
Re:I spent a day at a robot "carnival" in Shanghai. Here's what I saw
on Tuesday August 25, 2026 @07:02AM (#49431827)
Show HN: Noswoosh – instant macOS Space switching, no animation
Noswoosh: Instant, animation-free switching between macOS Spaces
Zero-Click Remote Code Execution: Exploiting MCP and Agentic IDEs
Gandalf: Agent Breaker (prompt injection game)
Re:Ask HN: Who is hiring? (February 2025)
on Tuesday February 04, 2025 @10:59AM (#42934061)
Lakera AI (https://lakera.ai) | Full-time | Zurich, Switzerland | ONSITE
We're looking for Software, Infrastructure, Platform, and ML Engineers (Senior-Staff level) in our office in Zurich.
Our team works on AI security (our flagship product is a layer for GenAI threat detection). Our current customers include Dropbox, Twilio and other large F500 companies.
We've created the game Gandalf, which helped us collect 20M+ attacks that we use to train our models.
If you're interested, it's best to reach out to me at [email protected] + apply at https://lakera.ai/careers!
Phishing with Gmail's Gemini Summarize
Phishing with Gmail's Gemini Summarize via prompt injection
Gandalf the Red: Adaptive Security for LLMs
The Beginner's Guide to Visual Prompt Injections
2024 Number Challenge
Studying LLM Generalization with Influence Functions [video]
The Beginner's Guide to Visual Prompt Injections
Show HN: Lakera – ChatGPT Data Leak Protection
Lakera – ChatGPT Data Leak Protection
Ready Player 50: Gandalf x CS50
Re:Rarbg Is No More
on Wednesday May 31, 2023 @11:00AM (#36139153)
Let's hope the same doesn't happen to Rutracker.
Re:Game to make an LLM reveal a secret password
on Friday May 12, 2023 @05:08AM (#35913739)
Try again! We increased the limits :)
Re:Game to make an LLM reveal a secret password
on Friday May 12, 2023 @05:06AM (#35913725)
That's interesting indeed. Could be that the context window of the model is too small...
Re:Game to make an LLM reveal a secret password
on Friday May 12, 2023 @03:05AM (#35912950)
We have deployed improvements that should help with throttling :)
Show HN: Gandalf - LLM Prompt Injection Challenge
Re:Ask HN: Who wants to be hired? (March 2023)
on Wednesday March 01, 2023 @12:47PM (#34985575)
Location: Zurich, Switzerland
Remote: Yes
Willing to relocate: No
Technologies: MLOps
Résumé/CV: https://www.linkedin.com/in/max-mathys/
Email: [email protected]
Re:$100B later, autonomous vehicles are still a car-wreck
on Saturday October 15, 2022 @06:13AM (#33213417)
Learning to Love Myself
Re:BMW Makes Heated Seats an $18/Month Subscription Service – Again
on Tuesday July 12, 2022 @06:09AM (#32067771)
Clickbait title. Should be renamed.
Re:A new hash algorithm for Git
on Tuesday February 04, 2020 @04:24AM (#22233760)
Just in case: http://archive.is/omsjJ