Starlog — Page 55
// LATEST
LLM Engineering
SGLang: How RadixAttention and Prefix Caching Achieve 5x Faster LLM Inference
LLM Engineering
Medusa: Accelerating LLM Inference by Predicting Multiple Tokens in Parallel
LLM Engineering
mergekit: Combining LLMs Without Training by Operating in Weight Space
AI Dev Tools
Promptmap: Testing LLM Security with a Judge, Jury, and Executioner Architecture
AI Dev Tools
PIPE: A Framework for When Prompt Injection Actually Matters
Data & Knowledge
Building Knowledge Graphs from Conversations: Inside MindGraph's Schema-Driven Architecture
Developer Tools
Building a Voice Interface That Runs Anywhere: Inside Open Interpreter's 01 Project
Developer Tools
Codel: Building a Self-Hosted Autonomous AI Agent with Docker Sandboxing
Developer Tools
OpenHands: The Open-Source AI Agent That Scored 77.6% on SWEBench
Automation
Untangle: Multi-Layer Web Server Fingerprinting Based on NDSS 2024 Research
Cybersecurity
dsieve: Taming Subdomain Chaos with Python-Style Slicing for Domain Lists
Developer Tools
Burpa: Turning Burp Suite Into a CI/CD Security Scanner
AI Agents
ZenGuard: Runtime Security for AI Agents That Actually Ships
AI Dev Tools
ps-fuzz: Testing GenAI Security with LLM-Against-LLM Combat
LLM Engineering
Building a Prompt Injection Firewall: Inside Vigil's Multi-Layer Defense System
Cybersecurity
NB Defense: Security Scanning for Jupyter Notebooks Before Your Secrets Hit GitHub
Cybersecurity
Mapping the Attack Surface: A Security Researcher's Guide to Foundation Model Vulnerabilities
LLM Engineering
How GPT-2 Leaks Its Training Data: A Deep Dive into Language Model Memorization
AI Agents
LiteLLM: The AI Gateway That Lets You Treat 100+ LLM Providers Like OpenAI
LLM Engineering
Inside Microsoft's Responsible AI Toolbox: A Widget Architecture for Model Accountability
AI Agents
DICES Dataset: When AI Safety Ratings Reveal More About Us Than The AI
Developer Tools
Inside METR's Task Standard: How AI Safety Researchers Benchmark Dangerous Autonomous Capabilities
Cybersecurity
Building an LLM Red Team Pipeline with Agentic Security
LLM Engineering