Production RAG at Scale: Lessons from Processing 10,000+ Listings Daily
A practical deep-dive into building a production RAG pipeline that scores thousands of documents daily with LLMs. Chunking, vector stores, cost optimization, and observability.
Tutorials
What makes an AI agent's decisions reliable
In the audits I have run, including this site's own, one thing keeps surfacing. An agent that is...
Which instruction files actually apply to your coding agent?
Deterministically trace the Codex and Claude Code repository instructions that apply to a target file.
PermitLint: audit Claude Code and Codex permissions locally
Permission files for coding agents can look restrictive while still allowing much more than intended....
I red-teamed my own LLM security gateway in four passes. Here's every gap I found.
I build a security gateway for LLM traffic. It sits in front of the model as a transparent...
What Makes Healthcare App Testing Different from Every Other App Category
A patient opens their prescription and sees 500mg instead of 50mg. A lab report displays "Normal"...
A Finish Notification Is a Lease, Not a Boolean
A coding agent finishes a turn. The monitor schedules an alert. Two seconds later, the developer has...
Spent the last 2 months building a #RAG engine for cognitive bias detection. Three assumptions completely fell apart: • More knowledge better retrieval • Passing evals production ready • More context better LLM output
Building a RAG Engine: Three Engineering Assumptions I Had to Unlearn ...
How to Give AI Agents Website Screenshots with MCP (2026)
Originally published on the Site-Shot blog. If you're building an AI agent that needs to see a web...
MCP Design Patterns: 6 Architectures for Your AI Tools
The Model Context Protocol is everywhere right now. Everyone is building MCP servers, connecting...
FullAgenticStack: Semantic Behavior Type: LinearAutoDestroy
Existe uma classe inteira de bugs e falhas de segurança que nasce de uma suposição errada: a de que o...
I built the first OSS federated threat-intel feed for LLM prompt injection. Here's how it works — and why no one else did.
Antivirus-style signature updates for LLM prompt injection. Ed25519-signed, pure-Python verification, Apache 2.0. 56 signatures live. Why no commercial defense tool ships this.
I built a tiny LLM circuit breaker: when the budget runs out, it fails over to a local model instead of failing or overspending
I run a small multi-agent system at home — a few agents that scout, summarize, and write, on a...
Xiaomi-Robotics-U0: A 38-Billion-Parameter Model for Unified Embodied Synthesis
What Changed Traditional foundation models for image and video generation, while powerful...
Latent-Identity Tuning: Achieving Fine-Grained Facial Edits in Text-to-Image Models Without Retraining
What Changed Traditional text-to-image personalization and editing methods often struggle...
The Last JSONL Line Is Not a State Machine
A transcript ended with an assistant message. Is Claude Code finished? Not necessarily. That line...
Топ 5 ИИ сказали, что мой смарт-контракт идеален. Я заставил их работать вместе и они нашли 4 бага
Для надежного аудита смарт-контрактов Web3 и устранения слепых зон одиночных нейросетей применяется...
An AI Science Workbench Needs a Reproducibility Graph, Not Just Chat History
Claude Science's auditable-artifact promise suggests a concrete graph for tracking datasets, code, environments, figures, and reviewer findings.
Before Enabling Copilot Agent in JetBrains, Make a Three-Issue Exit Test
GitHub Copilot Agent is now available in JetBrains AI Assistant. Evaluate the workflow with three real issues and portable evidence before adopting it.
My PM Brain Runs on Fear: The Loops I Built Around It
An update to https://dev.to/praneetaprakash/how-i-product-aec My PM brain is running in...
The Ultimate AI Coding Guide: Cursor vs. Claude Code vs. GitHub Copilot vs. Codex vs. Nim Code
The landscape of software engineering has shifted. We have moved past simple single-line...
Codex Needed Better Subagent Control. So I Built It.
I use Codex every day, and subagents are one of those features where the potential is immediately...
Building a Windows Laptop Monitoring Agent with Webex and Email Alerts
Building a Windows Laptop Monitoring Agent with Webex and Email Alerts I wanted to build a...
Your AI Agent Doesn't Have a Code Problem. It Has a Specs Problem.
I've been working over 15 years with clients, written specifications with them, and let's just say...