Blog

Ideas, insights, and everything we're building at TRAE.

Ship AI Software Faster with Vercel

Ship AI Software Faster with Vercel

How to build and deploy AI software in a fast way with TRAE and Vercel

Gary Qi
Best Practices for Agent Skills in TRAE

Best Practices for Agent Skills in TRAE

We've put together a guide on how to create, import, and use Agent Skills in TRAE.

Amber Shen
Build Backend Seamlessly with Supabase

Build Backend Seamlessly with Supabase

How TRAE integrates with Supabase.

Gary Qi
Making AI Coding Safer: How TRAE Protects Your Codebase

Making AI Coding Safer: How TRAE Protects Your Codebase

We're introducing sandboxing and a multi-layer security system designed to reduce security risk while keeping AI productive.

The TRAE Team
Cue-Pro Technical Deep Dive

Cue-Pro Technical Deep Dive

A technical deep dive into how we think and build Cue-Pro.

Bo Jiang
Introducing Cue-Pro: a bird's-eye view of your edit predictions

Introducing Cue-Pro: a bird's-eye view of your edit predictions

We upgraded Cue with repository-level edit sequences that keep you in flow

Amber Shen
TRAE 2025 Unwrapped

TRAE 2025 Unwrapped

This is your year in review. Let's look back at what we created together.

The TRAE Team
Quickstart to SOLO Coder

Quickstart to SOLO Coder

This guide will get you started with using a coding agent like SOLO coder in just a few minutes. We'll show you how to use SOLO Coder for some common development tasks.

Amber Shen
TRAE SOLO GA Release

TRAE SOLO GA Release

TRAE SOLO is the responsive coding agent we built for real-world software development.

Amber Shen
Scaling Velocity: How Keywords AI Accelerates Development with TRAE SOLO

Scaling Velocity: How Keywords AI Accelerates Development with TRAE SOLO

How a lean team builds production features end-to-end with an AI engineer that “understands, executes, and delivers.”

Hendrix Liu, Amber Shen
Max Mode in TRAE

Max Mode in TRAE

We are introducing Max Mode, which is designed for deep coding with larger context windows and more tool calls.

The TRAE Team
Best Practices for TRAE Rules

Best Practices for TRAE Rules

Stop repeating yourself. In this article we dive into how to set up TRAE Rules that remember your preferences, keep your team aligned, and avoid the common mistakes that make AI ignore your instructions.

Yunfeng Guan
Challenges and Considerations Behind Trae Cue

Challenges and Considerations Behind Trae Cue

Cue is one of the latest and most complex AI development tool features developed by TRAE. We explored the technical challenges, achievements as well as future plans for TRAE Cue.

Bo Jiang
Understanding Trae’s Data Practices

Understanding Trae’s Data Practices

This blog answers some of the most common questions about Trae’s data practices. We’ll explain where data is stored, how it’s used, and what options you have to control its collection and use.

The TRAE Team
Cue Major Update

Cue Major Update

Trae Cue gets its biggest monthly update with major improvements to model performance and speed.

Bo Jiang
Fresh New Look

Fresh New Look

TRAE 2.0 marks a full - scale design upgrade for a more focused, intelligent development experience.

Guangxi Cai
SOLO is All You Need

SOLO is All You Need

The AI that builds complete software.

Amber Shen
Trae Agent reached #1 on SWE-bench Verified, with Claude 4 - and it's open-sourced.

Trae Agent reached #1 on SWE-bench Verified, with Claude 4 - and it's open-sourced.

Trae Agent reached #1 on SWE-bench Verified, with Claude 4 — and it’s open-sourced.

Chao Peng, Pengfei Gao, Zhao Tian
Trae Agent 2.0: Smarter Architecture, Tools, and Memory

Trae Agent 2.0: Smarter Architecture, Tools, and Memory

Trae Agent 2.0 featured a reengineered architecture, unified agent modes, smarter tool orchestration, and long-term memory that brings us closer to building the real AI engineer.

Jiayu Tang
Mastering Context in Trae

Mastering Context in Trae

We've put together this guide that shows you how to make smart use of context in Trae to collaborate seamlessly with your AI partner.

Amber Shen
Trae Achieves #1 on SWE-bench Verified with Claude 3.7

Trae Achieves #1 on SWE-bench Verified with Claude 3.7

Trae achieves 71.0% accuracy on SWE-bench Verified with Claude 3.7, pushing the boundaries of automated issue resolving. Not just fast. Not just big. But deeply aligned with the nature of engineering.

Pengfei Gao, Zhao Tian, Chao Peng
More Agentic

More Agentic

We've improved our agent capabilities with better context understanding and greater autonomy.

Yang Shi
@Agent: Rules + Tools to Define Your Future Agents

@Agent: Rules + Tools to Define Your Future Agents

Deep dive into @Agent and how we built it.

Yang Shi
Collaborate with Intelligence

Collaborate with Intelligence

Trae unveils major updates: a unified Chat - Builder interface, @Agent system with MCP for multi - agent collaboration, and smarter context handling (#Web, #Doc, Rules). Boost your workflow with AI - driven coding and adaptable agents.

Yang Shi
Beyond AI Coding: Trae's Vision for Human-AI Collaboration

Beyond AI Coding: Trae's Vision for Human-AI Collaboration

Trae's vision is to empower seamless collaboration between humans and AI by advancing AI-driven coding, enabling developers to work smarter, faster, and more creatively.

Yang Shi