Fair on Average, Unfair in Practice: Why AI Governance Needs Simpson’s Paradox
Everyone in AI wants to talk about fairness.Continue reading on Medium »
ai, ai-ethics, ai-governance, Artificial Intelligence, genai
genai
ai, genai, gpt, llm, static-embeddings
Tokens & Embeddings: How LLMs Turn Text into Numbers (Explained So a First-Timer Actually Gets It)
You type a perfectly normal sentence into ChatGPT:“My app crashed after the update.”And the model replies like a human. It feels like the model is reading your words the way you do. It’s not. Under the hood, the model does something much more “engineer…
ai, genai, llm, prompt-injection-attack, Security
Blackwall LLM Shield -Because “Hope It Doesn’t Jailbreak” Isn’t a Security Strategy
Posted by Vish · Open Source · AI SecurityBlackwall-LLM-ShieldLet’s be honest. Most of us building AI products spend a lot of time thinking about prompts, models, latency, and costs. Security? That usually shows up as a last-minute checkbox — maybe a b…
agent-sandbox, Agentic AI, devops, genai, kubernetes
The AI Sandbox: Why Kubernetes Sandbox is the Future of AI Infrastructure
Beyond the StatefulSet Hack: A deep dive into SIG-Apps’ new native primitives for secure, 100ms-latency, and self-healing AI agent runtimes.If you give an AI Agent a terminal, you’ve given it a loaded gun. Kubernetes just built the safety lock.The land…
ai, embedding, genai, RAG, ranking, voyage-ai
Asymmetric Retrieval: Spend on Docs, Embed your Queries for Free
Documents are embedded once — worth the spend for maximum quality. Queries hit you on every request. This is what drives your cost at scale. Asymmetric retrieval with Voyage AI and Vespa. Real numbers, real config.
ai, genai, RAG, ranking, tensors
Build a High-Quality RAG App on Vespa Cloud in 15 Minutes
Retrieval-Augmented Generation (RAG) allows an LLM to answer questions using your data at query time. On their own, LLMs are powerful but limited: they can hallucinate, they have a fixed knowledge cutoff, and they know nothing about your private documents, internal wikis, or proprietary systems.