The Deska blog

Blog

Product notes, release highlights and how we build Deska: the one infinite canvas workspace for your editor, terminal, browser, notes and AI agents.

· 12 min read

The Agent as Study Partner for Hard Topics

Learn how to use an agent as study partner for hard topics like distributed systems or compiler theory using an infinite canvas and local-first AI tools.

· 11 min read

Using an Agent to Navigate Code You Do Not Know

Learn how to improve your developer workflow by using an agent to navigate code you do not know in complex repositories.

· 10 min read

The Agent as First Draft Machine

Learn why treating the AI agent as first draft machine improves developer velocity and how to manage large scale code generation within your workspace.

· 10 min read

Architecture Diagrams Generated From Code Reality

Explore how architecture diagrams generated from code can improve documentation workflows and how Deska integrates these visuals into an agentic workspace.

· 10 min read

Designing Approval Prompts Users Read

Learn technical strategies for designing approval prompts users read when building or using AI coding agents and automated developer tools.

· 11 min read

Unifying API Response Shapes Without Breaking Clients

Learn technical strategies for unifying API response shapes including versioning and adapters to maintain consistency without breaking existing client integrations.

· 10 min read

Generating Typed API Clients With an Agent

Learn how to use an AI agent to build and maintain typed API clients. Streamline your workflow by generating type-safe code using Deska side-by-side tools.

· 12 min read

Drafting Support Replies With Codebase Context

Learn how to improve technical support by drafting support replies with codebase context using AI agents and integrated development environments.

· 10 min read

Auditing Analytics Events Against Reality

Learn how to improve tracking hygiene by auditing analytics events against reality to ensure data integrity between code and your tracking plan.

· 10 min read

Scaffolding an Admin Panel You Will Not Hate

Learn modern strategies for scaffolding an admin panel that balances speed, flexibility, and developer experience without long-term technical debt.

· 10 min read

Agents That Install Packages You Did Not Ask For

Learn why AI agents install unrequested dependencies and how to manage agents that install packages you did not ask for through better policies and tools.

· 10 min read

The Agent Reached for any: Why AI Coding Models Fall Back on Lax Typing

Explore why the agent reached for any in your TypeScript code and how to fix lazy typing patterns in AI-generated pull requests.

· 10 min read

The Agent Added a Library: Check These Things

Learn what to verify when an AI coding agent adds a library to your project, from dependency bloat to security risks and license compatibility.

· 10 min read

Labels, Roles and Focus: An Agent A11y Sweep

Learn how to perform an agent a11y sweep by optimizing labels, roles, and focus management for AI coding agents and human users alike.

· 12 min read

Using Agents for Accessibility Fixes That Stick

Learn how to improve your workflow by using agents for accessibility fixes to automate remediation and maintain WCAG compliance in your codebase.

· 10 min read

White-Label Development With Agents in the Loop

Learn how to optimize white-label development with agents in the loop to scale your agency operations while maintaining code quality and security.

· 12 min read

The Agency of One: Agents as Your Staff

Learn how to build an agency of one leveraging AI agents to manage full-stack development, deployment, and monitoring from a local-first workspace.

· 10 min read

AFFiNE in 2026: Canvas Notes With a Familiar Pitch

A deep dive into AFFiNE in 2026 and how canvas notes are evolving for developers who need to organize complex documentation and visual workflows.

· 10 min read

Adversarial Verification: Agents Refuting Agents

Improve LLM reliability through adversarial verification. Learn how having agents refute agents on an infinite canvas leads to higher orchestration quality.

· 12 min read

Advent of Code With Agents: Cheating or Training?

Explore the ethics and technical benefits of using AI for Advent of Code With Agents while scaling your development skills through LLM orchestration.

💡 Ideas+🐛 BugsSuggest a feature or report a bug