In a recent LinkedIn post, Linas Beliūnas discusses a significant advancement in AI capabilities with Anthropic’s Claude AI, which he describes as effectively becoming a “24/7 employee that writes, reviews, and ships code while you sleep.” Beliūnas highlights the introduction of Claude Code Routines, a feature that transforms prompts into autonomous agents capable of operating within a code repository and executing tasks independently.
Beliūnas draws a parallel between this new functionality and the aspirations of earlier projects, stating:
“With this, Anthropic now has all the pieces of what OpenClaw promised but hosted, safer, and pointed at the highest-ROI task on earth: code.”
He elaborates on the practical implementation of these routines, emphasizing their ease of configuration and flexible triggering mechanisms. According to Beliūnas, users can set up these agents once, defining prompts, repository connections, and integrations with platforms like GitHub and Slack. The routines can then be activated on a schedule, via API calls, or in response to specific events such as a pull request being opened or an issue being created.
Autonomous Code Operations
The core innovation, as explained by Beliūnas, lies in the AI’s ability to spin up full Claude sessions independently, manage code edits, execute tests, and even open pull requests – all without direct human intervention or the need for local infrastructure.
A New Paradigm for Development Workflows
Beliūnas outlines a potential real-world workflow enabled by this technology:
“→ PR opens
→ Claude pulls issue context
→ Reviews diff, runs tests
→ Suggests fixes or ships a patch”
He underscores the transformative potential, noting that such tasks can be accomplished “All while you sleep 🤯.” This suggests a future where routine coding, testing, and review processes can be largely automated, freeing up human developers for more complex and strategic work.
From Concept to Deployable Reality
Linas Beliūnas points out that while open-source projects like OpenClaw have demonstrated the possibility of AI autonomy in coding, Anthropic’s approach makes this capability “boringly deployable.” This implies a shift from experimental concepts to practical, integrated solutions that businesses can readily adopt.
“Open-source agents like OpenClaw proved autonomy is possible. Anthropic just made it boringly deployable.”
He concludes by referring to this evolution as the arrival of “ClaudeBot 2.0,” signaling a new era in AI-assisted software development. Beliūnas also shared a personal anecdote, linking to a resource where he details how he transformed Claude into a “Personal COO that does work while I sleep,” further illustrating the practical application of these advanced AI agents.
📝 About This Content
This article is based on insights shared by Linas Beliūnas on LinkedIn.
📅 Originally posted on April 14, 2026 | View original post on LinkedIn →