Blog

Why LLM Inference Caches K and V but Never Q

#LLM#AI#Machine Learning#Inference#Transformers#Performance

Most engineers know that LLM inference uses KV cache. Far fewer can precisely explain why K and V are worth caching but Q is discarded after every decode step. The answer comes from the causal mask — and it has direct implications for how you size and optimize serving infrastructure....

5/29/2026

Why "Don't Store JWT in localStorage" — Auth History and the Cookie Return

#Security#Authentication#JWT#Web#Full Stack

The rule against storing JWTs in localStorage gets repeated as dogma, but most developers can't explain the full chain of reasoning — or why the industry is returning to HttpOnly cookies. This post traces the history from stateless JWT enthusiasm to the modern short-lived token + HttpOnly refresh cookie pattern....

5/27/2026

Always Set persist-credentials: false in actions/checkout

#GitHub Actions#Security#CI/CD#DevOps

The default persist-credentials: true in actions/checkout quietly leaves a GitHub token in a readable file for every subsequent job step. One setting closes the gap — here's why it matters and how to handle workflows that still need git auth....

5/25/2026

Microsoft Foundry Agent Service Is Becoming a Real Cloud Runtime

#Microsoft Foundry#Azure#Agents#Cloud#Enterprise AI

Microsoft's latest Foundry Agent Service docs read like a managed runtime story, not just another model-hosting layer. That matters for teams shipping enterprise agents into Azure....

4/29/2026

Automating Azure App Service Deployments with GitHub Actions and Secure Secrets

#CI/CD#GitHub Actions#Azure App Service#Secrets#DevOps#YAML

The Deployment Pipeline That Pays for Itself Manual deployments are a tax. Every manual step is a potential error, a context switch, and a reason deployments slow down as the team grows. CI/CD pipelin......

3/7/2026

GPT-5.4 Brings Stronger Coding, Tool Search, and Computer Use

#OpenAI#GPT-5.4#Coding Agents#Tool Search#Web Automation

GPT-5.4 folds frontier coding, better tool use, and computer interaction into one model. For teams building coding agents or browser automation, that changes the practical baseline....

3/5/2026

Dockerizing a Complex App: Lightweight Images, Multi-Stage Builds, and Secrets

#Docker#Containers#DevOps#Security#Multi-Stage Build#CI/CD

Why "Just Add a Dockerfile" Goes Wrong The first Dockerfile most developers write looks like this: dockerfile FROM python:3.12 WORKDIR /app COPY . . RUN pip install -r requirements.txt CMD "python", "......

2/20/2026

Next.js + FastAPI: The Full-Stack Combination That Replaced Our Monolith

#Next.js#FastAPI#Full Stack#Performance#Tailwind#DX

The Monolith We Were Running From Two years ago, a typical internal tool at Foetron was a Flask app serving Jinja2 templates, jQuery for interactivity, Bootstrap for styling, and a PostgreSQL database......

2/5/2026

Connecting Custom AI Agents to OpenWebUI: Auth, Latency, and API Design

#AI#OpenWebUI#API Design#Authentication#FastAPI#Integration

Why OpenWebUI as the Interface Layer OpenWebUI is an open-source, self-hostable chat interface that supports the OpenAI API spec. That compatibility means any system that implements /v1/chat/completio......

1/24/2026

SchemaForge: Extracting Structured Data from Unstructured Documents with Azure AI

#Azure AI#Document Intelligence#Computer Vision#OCR#Python#SchemaForge

The Problem with Unstructured Documents An enterprise generates thousands of documents per month: invoices, compliance reports, technical specifications, contracts, procurement forms. Each document ty......

1/8/2026

SQL Database Scheduled Backup to Azure Blob

#SQL Server#Azure#Backup#Automation

Automate backups of multiple SQL Server databases directly to Azure Blob Storage....

12/12/2025

Designing Agentic AI Workflows: Tool-Calling, Memory, and Multi-Modal Integration

#AI Agents#LLM#Tool Calling#Memory#Multi-Modal#VirtuAI

What Makes a System "Agentic"? An LLM that answers questions is not an agent. An agent is a system that can: 1. Perceive — receive input from the environment (text, images, API responses, files) 2. Pl......

12/12/2025

Building a Production RAG Pipeline with Azure AI Search and GPT-4

#AI#RAG#Azure AI Search#LLM#Vector Search#Python

What RAG Actually Solves Large Language Models hallucinate. They also have a knowledge cutoff. And they can't know about your internal documentation, your product catalogue, or your VM specification d......

11/29/2025

Building a Cloud Cost Calculator with the Azure Retail Prices API

#Azure#Cost Optimization#Pricing API#Cloud Economics#Architecture

Why First-Party Pricing Calculators Fall Short The Azure Pricing Calculator is excellent for static estimates. It falls short the moment you need to answer questions like: "If we move from 8 D4s_v3 no......

11/15/2025

Private Endpoints on Azure: The DNS Gotchas Nobody Warns You About

#Azure#Networking#DNS#Private Endpoints#VNet

Why Private Endpoints Break Your DNS (And How to Fix It) Private Endpoints are one of Azure's most powerful security features — they pull a PaaS service (like Azure SQL, Storage, or Key Vault) into yo......

10/28/2025

Securing Azure Virtual Desktop with Entra ID and Passwordless MFA

#Azure#AVD#Entra ID#Security#MFA#Passwordless

The Problem with Passwords in VDI Environments Traditional Virtual Desktop Infrastructure relies on Active Directory and password-based authentication. In a cloud-first world, that model has three cri......

10/3/2025

Monolith to Microservices on Azure: A Production Migration Playbook

#Azure#Architecture#Microservices#Migration#Cloud

Why the Monolith Had to Go Every engineering org reaches the same inflection point: the codebase that once shipped features in a weekend now requires a war room to deploy. That was the reality at Foet......

9/12/2025

AI SDK 5 Pushes Type-Safe Agents Further Into the Frontend

#Vercel#AI SDK#TypeScript#React#Agents#Frontend

Vercel's AI SDK 5 is a meaningful web-dev release because it tightens the contract between model output, tool state, and UI rendering instead of leaving frontend teams to improvise it....

7/31/2025

OpenAI's Responses API Is the New Default for Agentic Web Apps

#OpenAI#Responses API#Agents#Web Search#File Search

The Responses API collapses chat plus tool orchestration into a single primitive. That simplifies how teams add web search, file search, and agent execution to real applications....

3/11/2025