• Pricing
  • Customers
  • Docs

For Mastra Agents

For Any Agent

OS FrameworkTS First agentic framework
StudioObserve, evaluate, collaborate
ServerCloud deployment for agents
Memory GatewayAI Gateway with Memory

Get Started

Latest

Books & Podcast

TemplatesLearn from examplesLearn • MCP CourseAgent base CLI courseLearn • Video CourseVideo course on agents
BlogLatest from Mastra teamResearchResearch and benchmarksChangelogUpdates and Improvements
Book • PrinciplesPrinciples of Building AI AgentsBook • PatternsPatterns of Building AI AgentsPodcast • Agent HourWeekly livestreams and guests
  • 22.9k
  • LoginSign Up
  • product
    • OS FrameworkTS First agentic framework
    • StudioObserve, evaluate, collaborate
    • ServerCloud deployment for agents
    • Memory GatewayAI Gateway with Memory
  • resources
    Get Started
    • TemplatesLearn from examples
    • Learn • MCP CourseAgent base CLI course
    • Learn • Video CourseVideo course on agents
    Latest
    • BlogLatest from Mastra team
    • ResearchResearch and benchmarks
    • ChangelogUpdates and Improvements
    Books & Podcast
    • Book • PrinciplesPrinciples of Building AI Agents
    • Book • PatternsPatterns of Building AI Agents
    • Podcast • Agent HourWeekly livestreams and guests
  • more
    • Pricing
    • Customers
    • Docs
    • Contact
  • 22.9k
  • LoginSign Up
Daniel Lew

Daniel Lew

Software Engineer

Daniel Lew is a software engineer at Mastra based in Waterloo. A University of Waterloo graduate with a background in psychology, he previously worked at Netlify and Gatsby.

3 blog posts

Posts by Daniel Lew

Announcing Agent Editor

Edit agent instructions, tools, and display conditions from Studio with draft/publish versioning, or programmatically through the editor API.

Apr 8, 2026
Daniel Lew

Building low-latency guardrails to secure your agents

How we built a suite of out-of-the-box input processors and optimized them from 5000ms to under 500ms per request.

Jul 30, 2025
Daniel Lew

Reducing tool calling error rates from 15% to 3% for OpenAI, Anthropic, and Google Gemini models

We recently built a tool compatibility layer that reduced tool calling error rates from 15% to 3% for 12 OpenAI, Anthropic, and Google Gemini models, across a set of 30 property types and constraints.

May 29, 2025
Daniel Lew
Mastra Logo

Product

FrameworkStudioServerGateway

Framework

AgentsRAGWorkflowsObservability

Developers

DocsChangelogTemplatesLearn · MCP CourseLearn · Video Course

Resources

BlogResearchBook · PrinciplesBook · PatternsPodcast · Agent Hour

Company

AboutCustomersCareersContact UsNewsletter

Connect

Contact UsGitHubDiscordYouTubeX (Twitter)
SecurityFully operational
Privacyllms.txt