Applied research, in the open.
Comparisons, teardowns, and working notes from the practice — published as they’re useful.
Nobody tells the vibe coders. Production does.
A viral list of 107 infrastructure terms is making the rounds as a gotcha for AI-assisted builders. It's half right. Here's the half that matters, and the env var that humbled me this week.
ReadFinally, an Open Standard for the Karpathy LLM Wiki is HERE
This video introduces the Open Knowledge Format (OKF), a proposed standard for structuring personal knowledge bases built with LLMs. It addresses the lack of interoperability between different LLM wiki implementations, e
ReadQwen-AgentWorld: A New Paradigm in AI Agent Training with World Models
This video explores the Qwen-AgentWorld, a world model that enables AI agents to simulate environments and predict outcomes of actions, improving their reasoning and robustness. It's a major step in training AI agents fo
ReadNext.js 15's New Caching API: A Deep Dive into `use cache` and Dynamic IO
Next.js 15 introduces a new caching API called use cache and Dynamic IO, which simplifies caching and dynamic rendering in server components. This is particularly useful for developers looking to optimize performance and
ReadAI Agents for DevOps: Understanding, Building, and Optimizing
This video explains AI agents from a DevOps and platform engineering perspective, covering their structure, context window limitations, and optimization techniques like sub-agents and agent skills. It's particularly rele
ReadMCP Explained: The New AI Standard You Need to Know
This video explains MCP (Model Context Protocol), an open-source standard for connecting AI models to external systems like Kubernetes, Git, and databases. It simplifies AI-agent interactions with infrastructure by repla
ReadSecure AI Traffic with AgentGateway: A Deep Dive into MCP Security
AgentGateway is a high-performance gateway for managing and securing MCP (Model Control Protocol) traffic in AI and agent systems. It enables centralized control over how AI applications interact with external systems li
ReadMCP Gateway on Kubernetes Using Agent Gateway
This video explains how to deploy and configure Agent Gateway on Kubernetes to route traffic to MCP servers, which are tools used by large language models (LLMs) to interact with systems like Kubernetes. It covers the se
ReadSecure AI Traffic with Agent Gateway and Kubernetes
This video introduces Agent Gateway, a Kubernetes-based proxy that routes AI traffic to different LLM providers or local models using the Gateway API. It's a powerful tool for managing AI traffic in a scalable, secure, a
ReadTalk to Claude on 3CX Phone System Tutorial (Full Setup)
This video demonstrates how to set up a phone system using 3CX and integrate it with Claude for voice-based interactions. It covers deploying a cloud-based phone system, configuring a Raspberry Pi as a voice server, and
ReadThe Hidden Dangers of Malware and IoT Security: Insights from a Threat Research Director
Bogna Botato, a threat research director at Bit Defender, discusses the evolving landscape of malware, the dangers of advanced persistent threats (APTs), and the growing risks of insecure IoT devices. She highlights the
ReadInside the Secretive Draco Team: Combating Cybercrime with AI and Law Enforcement
This interview with Alex Kosawe of Bit Defender reveals the inner workings of the Draco team, a secretive group that collaborates with law enforcement to investigate and disrupt cybercrime. The team uses AI and automatio
ReadAI Agents Are the New SaaS: A Playbook for Building Agent-First Startups
This video outlines a strategic framework for building agent-first startups, positioning AI agents as the next evolution of SaaS. It emphasizes identifying repetitive, labor-intensive workflows, automating them with AI,
ReadThe Best AI Coding Setup Isn't the Most Autonomous One (Here's Why)
This video explores the five levels of AI coding autonomy, from manual coding with AI as a reference tool to fully autonomous "dark factories" that ship code without human intervention. It emphasizes that the most reliab
ReadMiniCPM-5: A 1B Cognitive Core for On-Device Intelligence
This video explores MiniCPM-5, a 1B parameter model from OpenBMB, as a candidate for the "cognitive core" concept — a small, reasoning-capable model that can run on-device and use tools. It discusses its architecture, tr
ReadThe Six Most Valuable Skills in the Agentic Era
This video identifies six core skills that will be increasingly valuable as AI becomes more pervasive. These include managing AI agents, building distribution, robotics engineering, content curation, building and distrib
ReadAI Sovereignty, Data Control, and the Future of Work: Key Insights from the All-In Podcast
This podcast discusses the growing tension around AI sovereignty, the risks of data commoditization, and the implications of AI on the job market. It also touches on political and fiscal issues in California, including b
ReadPoisoning Data for Privacy: How Cloaked is Reshaping Digital Identity
Cloaked, a privacy-focused startup, has raised over $400 million to develop tools that "poison" personal data to protect users from tracking and exploitation. The company offers services like email masking, credit card g
ReadThe Only 19 Skills to Make Money in the AI Era
Dan Martell ranks 19 skills based on their barrier of entry, defensibility against AI, and profitability. The video highlights skills like public speaking, trades, and AI automation as high-value, while others like SEO a
Read