Shane Thomas
CPO
Shane Thomas is the founder and CPO of Mastra. He co-hosts AI Agents Hour, a weekly show covering news and topics around AI agents. Previously, he was in product and engineering at Netlify and Gatsby. He created the first course as an MCP server and is kind of a musician.
Posts by Shane Thomas
How to build AI agent evaluation that ships reliable agents
Choose metrics, build graders, run evals in CI/CD, and monitor production to catch failures before your users do.
Aron SchuhmannEvalsJun 22, 2026AI gateway: one integration point for every provider
Learn what an AI gateway does, how it routes requests across providers, and how to implement one for cost control, observability, and reliability.
Aron SchuhmannModels & RoutingJun 22, 2026RAG platforms: what they are, how they work, and how to choose one
Learn what a RAG platform does, how RAG pipelines work end to end, and what to evaluate when choosing one for production AI agents.
Aron SchuhmannRAGMar 23, 2026Mastra Changelog 2026-03-23
Token-aware model routing for observational memory, MongoDB-backed versioned datasets and experiments, Okta SSO auth with RBAC, workflow upgrades, tool suspension handling, and observability improvements.
Shane ThomasChangelogMar 17, 2026Mastra Changelog 2026-03-17
AI Gateway tool support in the agentic loop, more reliable observational memory, MCP client diagnostics and per-server control, and improved tracing with spanId, working memory and schema fixes.
Shane ThomasChangelogMar 13, 2026Mastra Changelog 2026-03-13
Persistent agent workspace filesystem, observability pipeline upgrades, type-safe server route inference, lower token costs for long-running Observational Memory, and breaking changes to metrics, scores, and ObservabilityBus configuration
Shane ThomasChangelogMar 12, 2026Mastra Changelog 2026-03-12
Cloudflare Durable Objects storage adapter,improved observability for MCP tooling, more reliable agent loops and token budgeting, sandbox and workspace extensibility via provider-specific getters and string PIDs
Shane ThomasChangelogMar 11, 2026Mastra Changelog 2026-03-11
Dynamic model fallback arrays, Standard Schema normalization with Zod v4 compatibility, customizable request validation for server adapters, and end-to-end RequestContext support for tracing and datasets/experiments.
Shane ThomasChangelogMar 5, 2026Mastra Changelog 2026-03-05
Improve tool-call accuracy, MCP fetch hooks receive RequestContext, and reliability updates surface streaming errors, preserve history in stateless tool runs, and clean up orphaned vector embeddings.
Shane ThomasChangelogMar 4, 2026Mastra Changelog 2026-03-04
Token-aware workspace output, smarter truncation, LSP resolution, sandbox process control, end-to-end auth and RBAC, workflow path tracking, concurrent-safe snapshots, and harness.sendMessage() file support.
Shane ThomasChangelogFeb 26, 2026Mastra Changelog 2026-02-26
Supervisor pattern for multi-agent coordination, metadata-only vector queries, more flexible runEvals options, LSP diagnostics after workspace edits, and a new Blaxel sandbox provider.
Shane ThomasChangelogFeb 24, 2026Mastra Changelog 2026-02-24
Background process management in workspaces and sandboxes, runtime tool configuration via Workspace.setToolsConfig(), and observational memory reliability and introspection improvements.
Shane ThomasChangelogFeb 23, 2026Mastra Changelog 2026-02-23
AST-based workspace edits, real-time streaming tool argument previews, built-in task tracking tools for Harness, and improved Observational Memory continuity
Shane ThomasChangelogFeb 19, 2026Mastra Changelog 2026-02-19
A new reusable Harness orchestration layer for agent apps, versioned workspaces and skills, pluggable blob storage, security and discovery upgrades, plus improved tool I/O and streaming behavior, including better chunk handling.
Shane ThomasChangelogFeb 13, 2026Mastra Changelog 2026-02-13
Datasets and experiments for evaluations, workspace and filesystem lifecycle improvements with safer introspection, and workflow foreach progress streaming.
Shane ThomasChangelogFeb 11, 2026Mastra Changelog 2026-02-11
Observational memory async buffering (default-on) with new structured streaming events, workspace mounts with CompositeFilesystem for unified multi-provider file access.
Shane ThomasChangelogFeb 11, 2026Mastra Changelog 2026-02-11
Observational memory async buffering (default-on) with new structured streaming events, workspace mounts with CompositeFilesystem for unified multi-provider file access.
Shane ThomasChangelogFeb 4, 2026Mastra Changelog 2026-02-04
Observational Memory, skills.sh integrations, plus improved tracing and safer error handling.
Shane ThomasChangelogJan 30, 2026Mastra Changelog 2026-01-30
Unified Workspace API, observability and server adapters improvements.
Shane ThomasChangelogJan 20, 2026Mastra Changelog 2026-01-20
Server adapters, composite storage, processor system overhaul, AI SDK v6 support, thread cloning, and more.
Shane ThomasChangelogNov 20, 2025Mastra Changelog 2025-11-20
Generate endpoint fix for OpenAI streaming, AI SDK v5 fine-grained stream control, and more.
Shane ThomasChangelogNov 14, 2025Mastra Changelog 2025-11-14
Mastra v1 is now in beta, we published a new book, and shipped the Mastra UI Dojo.
Shane ThomasChangelogNov 1, 2025Mastra Changelog 2025-11-01
Agent Studio, 1.0 Release Preparation, and more.
Shane ThomasChangelogOct 28, 2025HITL: Ask Before Acting
Learn how to use HITL (Human-in-the-Loop) to safely build tools that require human approval.
Shane ThomasWorkflowsOct 23, 2025Mastra Changelog 2025-10-23
Workflow state management, unified streaming, AI tracing updates, and more..
Shane ThomasChangelogOct 3, 2025Mastra Changelog 2025-10-03
New model router and automatic model fallbacks.
Shane ThomasChangelogSep 25, 2025Mastra Changelog 2025-09-25
New streaming architecture, scoring improvements, and more.
Shane ThomasChangelogSep 18, 2025Mastra Changelog 2025-09-18
Resumable streams, structured output API change, bundling overhaul, and more.
Shane ThomasChangelogSep 8, 2025Mastra Changelog 2025-09-08
Zod v4 support with v3 compatibility and more.
Shane ThomasChangelogAug 21, 2025Mastra Changelog 2025-08-21
New streamVNext and generateVNext methods with AI SDK v5 support, output processors, and more.
Shane ThomasChangelogAug 19, 2025MASTRA.BUILD Templates Hackathon Winners & Community Templates Library
Celebrating the winners of our MASTRA.BUILD templates hackathon and launching our new community templates library
Shane ThomasCompanyAug 14, 2025Mastra Changelog 2025-08-14
This week we've shipped a major RAG enhancement with semantic markdown chunking and updated our A2A implementation to spec v0.3.0.
Shane ThomasChangelogAug 8, 2025Mastra Changelog 2025-08-08
Introducing Scorer API, critical fixes for message handling and parallel workflows, plus improvements to memory filtering and type safety across the board. Plus, we're announcing our first ever conference: TypeScript AI
Shane ThomasChangelogJul 30, 2025Mastra Changelog 2025-07-30
Improvements to Agents, Memory, and more.
Shane ThomasChangelogJul 24, 2025Mastra Changelog 2025-07-24
Introducing Mastra Templates, MASTRA.BUILD hackathon, StreamVNext protocol, complete eval scores system, and major improvements across the board.
Shane ThomasChangelogJul 23, 2025Introducing Mastra Templates
Announcing Mastra Templates: pre-built, production-ready, customizable agent and workflow projects. Plus: join our MASTRA.BUILD hackathon to create and showcase your own templates.
Shane ThomasDeploymentJul 17, 2025Mastra Changelog 2025-07-17
New memory improvements, new CLI templates, reasoning display in playground, and major improvements across the board.
Shane ThomasChangelogJul 9, 2025Mastra Changelog 2025-07-09
Mastra is now Apache-2.0 licensed, Playground goes multi-modal, new memory and RAG features, and more.
Shane ThomasChangelogJul 3, 2025Mastra Changelog 2025-07-03
Agent Network (vNext), workflow cancellation, and custom memory model support highlight this week's Mastra updates.
Shane ThomasChangelogJun 27, 2025Mastra Changelog 2025-06-27
Mastra Cloud public beta, agent network chat, memory improvements, workflow updates, and a new Mastra 101 lesson.
Shane ThomasChangelogJun 26, 2025Mastra Cloud Public Beta
Mastra Cloud is now in public beta — deploy, manage, and scale your AI agents and workflows.
Shane ThomasDeploymentJun 20, 2025Mastra Changelog 2025-06-20
We added sleep methods to workflows, structured memory to agents, and a new Gladia STT provider.
Shane ThomasChangelogJun 13, 2025Mastra Changelog 2025-06-13
Cross-thread memory recall, universal schema support, and enhanced workflow observability.
Shane ThomasChangelogJun 6, 2025Mastra Changelog 2025-06-06
Mastra 101, Mastra Auth, and more
Shane ThomasChangelogJun 4, 2025Introducing Mastra 101: The experimental course led by MCP, right in your IDE
The newest (and perhaps coolest) way to learn Mastra
Shane ThomasAgentsMay 29, 2025Mastra Changelog 2025-05-29
MCP Server updates, streaming support, and more.
Shane ThomasChangelogMay 27, 2025MASTRA.BUILD Recap & Winner Announcement
The results of MASTRA.BUILD, our inaugural hackathon.
Shane ThomasCompanyMay 23, 2025Full-Stack TypeScript Agents with Mastra and Copilot Kit
Build a full-stack TypeScript agent with Mastra and Copilot Kit.
Shane ThomasUse CasesMay 15, 2025Mastra Changelog 2025-05-15
Mastra's latest updates: A2A support, MCP updates, and more.
Shane ThomasChangelogMay 10, 2025Building an Interactive Detective Game with Multi-Agent AI
How we built an interactive detective game using Mastra's agent system and Next.js.
Shane ThomasMulti-agentMay 7, 2025Mastra Changelog 2025-05-07
Mastra's latest updates: vNext is live, AGUI integration, and linting.
Shane ThomasChangelogMay 7, 2025Introducing Mastra Docs Chatbot: Your AI Documentation Assistant
Explore how our new docs chatbot leverages the Mastra MCP Docs server to enhance user experience and streamline access to documentation.
Shane ThomasMCPMay 1, 2025Mastra Changelog 2025-05-01
Mastra's latest updates: vNext workflows, MongoDB vectorDB provider, Streamable HTTP MCP transport, and more.
Shane ThomasChangelogApr 24, 2025Mastra Changelog 2025-04-24
Mastra's latest updates: dynamic agents, MCPServer support, vNext workflows, and more.
Shane ThomasChangelogApr 23, 2025Building a Web Browsing Agent with Mastra and Stagehand
See how we built a web browsing agent using Mastra and Stagehand. Learn about the tools needed to allow your agents to control a browser.
Shane ThomasTools & IntegrationsApr 18, 2025Building a Personal Assistant with Mastra and MCP
How we built a personal assistant that can manage emails, monitor GitHub activity, schedule social media posts, and more - all through a Telegram interface.
Shane ThomasMCPApr 16, 2025Mastra Changelog 2025-04-16
Mastra's latest updates: MCP registry registry, model settings in Playground UI, and new Memory docs
Shane ThomasChangelogApr 11, 2025PDF-to-Blog: Giving documents a second life with Mastra and Mistral OCR
A technical guide showing how to build a PDF-to-blog converter using Mastra and Mistral OCR
Shane ThomasRAGApr 4, 2025Mastra Text-to-SQL: from natural language to database queries
Use Mastra's Text-to-SQL system to convert natural language to database queries
Shane ThomasDatabases & StorageApr 2, 2025Mastra Changelog 2025-04-02
Mastra's Latest Updates: API routing, memory processors, and more
Shane ThomasChangelogMar 27, 2025Mastra Changelog 2025-03-27
Mastra's Latest Updates: Nested Workflows, Voice to Voice, and more
Shane ThomasChangelogMar 4, 2025Mastra Changelog 2025-03-04
Introducing Mastra Cloud and Mastra Voice (as well as a few other things)
Shane ThomasChangelogFeb 28, 2025From Whiteboard to Excalidraw: Building a Multi-Agent Workflow
How we built a whiteboard image to Excalidraw converter using Mastra's multi-agent workflows, and what we learned about breaking down complex AI tasks into deterministic steps with validation loops.
Shane ThomasWorkflowsFeb 13, 2025Using Vercel's AI SDK with Mastra
How Mastra enhances Vercel's AI SDK with agent primitives, RAG pipelines, and evaluation capabilities
Shane ThomasModels & RoutingJan 25, 2025Every API needs a natural language endpoint
Building a world where every API has a chat endpoint
Shane ThomasTools & Integrations