Articles by Josh.
Redis Pricing Compared: Every Major Provider in 2026 (With Real Numbers)
Redis pricing compared across Upstash, AWS ElastiCache, MemoryDB, Redis Cloud, Google Memorystore, Azure, DigitalOcean, and Aiven, with real numbers and example cost scenarios.
Building Subagents in the Vercel AI SDK v6
How to build subagents in the AI SDK v6 with ToolLoopAgent, and how to share state between them using Upstash Redis.
Building extremely fast agent memory with virtual markdown
A virtual markdown file is the perfect interface for AI agent memory. No third-party tools needed, AI SDK v6 and Redis.
Upstash Redis vs Cloudflare KV: Features, Performance & Pricing in 2026
The 2026 Upstash Redis vs. Cloudflare KV comparison. Features, limitations, pricing and performance.
Free Redis databases for AI agents, in one HTTP request
POST https://upstash.com/start-redis returns a free Upstash Redis database. No signup and no API key needed. Perfect for AI agents.
Building a really fast virtual filesystem for AI agents
How to build a virtual filesystem on top of Upstash Redis, with Redis Search powering grep for AI agents.
Give AI SDK agents a computer: Filesystem, Shell and Git
How to run Vercel AI SDK agents isolated cloud sandboxes. Give an AI SDK agent filesystem, shell and git access.
A First Look at Upstash Redis Search
An early look at Upstash Redis Search. A type-safe, serverless way to search your data, built on Redis.
Introducing Dynamic Rate-Limits for Upstash Ratelimit
Upstash Ratelimit now supports dynamic rate limits. You can now change rate-limits without redeploying.
Upstash Prod Pack: Enterprise-Grade Redis for $200/month
Prod Pack adds 99.99% uptime SLA, SOC 2 compliance, dedicated support, and advanced monitoring to any paid Upstash Redis or QStash instance.
Context7 Without Context Bloat
Context7's new architecture decreases token usage by 65% and reduces latency by 38%.
Integrating Upstash MCP is a lot easier now
The Upstash MCP server setup just got simpler. One CLI command for Claude, Cursor, or Copilot.
Redis functions are here!
Upstash Redis SDK now supports Redis Functions. Like stored procedures for Redis, with full TypeScript support.
Figma-Style Live Cursors with Next.js 16 & Upstash Realtime
Build Figma-style live cursors in Next.js 16 with Upstash Realtime.
Realtime Notifications with Next.js 16 & Vercel
Build a realtime notification system in Next.js 16 with Upstash Realtime.
Resumable AI SDK v5 Streams with Upstash Realtime
Build AI chat streams that survive page refreshes and network disconnects using Vercel AI SDK v5, Upstash Workflow, and Upstash Realtime.
The Easiest Way to Add Realtime in Next.js
Introducing Upstash Realtime - a type-safe, serverless alternative to Pusher and Ably. Add live updates, cursors, and notifications to Next.js in 60 seconds.
Building Reliable & Type-Safe Webhooks
In this article, I wanna show you how easy it is to add a fully type-safe webhook system to your own app. We have automatic retries delivery guarantees delays ...and much more already built-in with…
Smooth Text Streaming in AI SDK v5
There's one pattern I've noticed over and over again in some of the most polished AI chat interfaces. This pattern is used by huge companies like Anthropic in production and make the AI chat…
Saving AI SDK v5 Chat Messages in Redis
I wrote an updated 2026 version of this article for Vercel AI SDK v6 over at xadd.dev! In 2023, Vercel introduced the first version of their AI SDK. In their words, it started as: "An interoperable,…
Building a seriously reliable serverless API
A few days ago, the internet broke. GCP, AWS, Cloudflare, nearly every major cloud provider had downtime: Most web apps (ofc mine included) rely on third-party APIs like OpenAI, an email service, or…
Making Next.js Navigation Fast af
Next.js has been my go-to framework for building full-stack apps for years. Built-in API routes, image optimization, and good integrations, there are a lot of reasons to use it. But there's one part…
F*ck Clean Code, You Need an Idea
There's a software development philosophy that aligns perfectly with my understanding of shipping ideas extremely fast. I've seen it work again and again over the past year. It helped me launch…
How to Build LLM Streams That Survive Reconnects, Refreshes, and Crashes
In this article, we're building extremely durable LLM streams that easily survive: Network outages Page refreshes Closing the website Closing the laptop lid Bonus: You can view the same stream on…
Context7 MCP: Up-to-Date Docs for Any Cursor Prompt
Last month, we introduced Context7 - a way to stop Cursor, Claude or any LLM from generating broken, outdated code. The idea: feed your LLM the latest documentation for the libraries you're using —…
Model Context Protocol (MCP): Build Your Own in 5 Minutes
The Model Context Protocol (MCP for short) is a standardized protocol to connect AI to external APIs, live data and more. It was introduced by Anthropic in late 2024. Think of an MCP server as a tool…
Building a Fast, Typo-Tolerant AI Search Engine
Recently, I wanted to build a production search functionality for a tech stack I recently released called JStack. While researching how to build this, I realized that building search is usually a…