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.
- How Agents Reduce Startup FrictionLearn how agents reduce startup friction in development by automating environment setup and context gathering for faster project launches.
- Running Coding Agents Overnight, SafelyLearn strategies for running coding agents overnight safely. Implement guardrails, logging, and remote monitoring for autonomous development sessions.
- Every Agent Against the Same Legacy RepoExplore how different AI coding agents perform against the same legacy repo when managed through a unified developer workspace.
- Every Agent on a Fresh Project: Evaluating AI Coding WorkflowsCompare how every agent on a fresh project handles initialization, architecture, and code generation using local-first tools and Deska.
- AGENTS.md: The Config File Every Repo Should Have NowLearn how an AGENTS.md guide helps AI tools understand your code. Master the cross-agent instruction file for better automated development workflows.
- Nightly Agent Jobs: Lint Debt Down While You SleepLearn how to use nightly agent jobs to automate linting and technical debt reduction using local tools and AI coding agents.
- Agents for Exploration, You for DecisionsLearn why the most effective developer workflows use agents for exploration while keeping humans in charge of final technical decisions.
- Agents on Docker Compose Projects: Containers as ContextLearn how to use AI coding agents on Docker Compose projects to automate development workflows and manage multi-container environments with context.
- Two Agents Changed the Same File: Solving LLM Edit ConflictsLearn how to manage scenarios where two agents changed the same file. Strategies for multi-agent coordination and preventing race conditions in AI coding.
- Momentum: Agents and the Motivation LoopLearn how agents and the motivation loop impact developer productivity and how Deska maintains momentum through local-first infinite canvas workspaces.
- Do Agents Make the Four Day Week Inevitable?Explore how AI coding agents and autonomous workflows impact developer productivity and whether they make the four day week inevitable for modern teams.
- Agentic RAG: When Retrieval Needs ReasoningExplore how Agentic RAG shifts from static retrieval to autonomous reasoning for complex developer workflows and multi-step information gathering.
- Agentic Loops: Observe, Act, Verify, RepeatMaster the concept of agentic loops in AI software engineering. Learn how observe, act, and verify cycles power modern coding agents like Claude Code and Codex.
- XSS Audits of Your Own App With an AgentLearn how to use AI coding agents to perform XSS audits of your own app, identifying reflected and DOM-based vulnerabilities through automated security flows.
- Let the Agent Write While You Review: A Two-Track WorkflowOptimize your output by letting the agent write while you review code in parallel. Learn how a two-track workflow increases developer throughput.
- Delegating the Boring Half of the Test SuiteLearn how to delegate the boring half of the test suite to AI agents while maintaining architectural control and code quality in your dev workflow.
- One Agent Writes Code, Another Writes TestsLearn how to improve software quality by having one agent writes code and another writes tests in a side-by-side multi-agent development workflow.
- Migrations Written by Agent, Reviewed by YouLearn how migrations written by agent workflows improve database reliability and how tools like Deska facilitate human review to prevent data loss.
- Regexes With Test Cases, Written TogetherLearn how to generate regexes with test cases using AI agents and local-first workspaces to ensure pattern reliability and security.
- Making Agents Write Documentation People ReadLearn how to optimize workflows when making agents write documentation to ensure clarity, accuracy, and maintainability for modern software teams.