How one Anthropic seller rebuilt his team's workflows with Claude Code

· Source: Claude Blog · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering, Robotics & Autonomous Systems · Depth: Intermediate, medium

Summary

Jared Sires, a GTM product manager at Anthropic, transformed his team's sales workflows by developing Claude-powered automation tools, despite having no prior coding experience before joining in 2024. As an account executive managing 600-700 accounts, Sires faced overwhelming email volumes and administrative tasks. He leveraged Claude Code and Apps Script to create CLAFTS (Claude Drafts), a Gmail-integrated application that drafts customer email replies, saving him 2-3 hours daily. This innovation quickly spread within the sales organization, leading to his current role focused on building Claude-based solutions. Sires further developed "daily brief" and "daily recap" skills, which utilize web search, Google Calendar, and CRM data to prepare for meetings and generate follow-up emails. These tools, packaged as a plugin within Claude Cowork, are now used by approximately 80% of Anthropic's sales team, integrating with platforms like Salesforce, Intercom, and Gmail, significantly streamlining onboarding for new hires.

Key takeaway

For sales professionals or GTM teams struggling with administrative overload, you should explore using large language models like Claude Code to automate repetitive tasks. By building custom tools for email drafting, pre-call research, and follow-up generation, you can reclaim 10-15 hours weekly. This allows you to focus on meaningful customer conversations and strategic sales activities, significantly improving efficiency and accuracy in your daily workflows.

Key insights

Non-technical professionals can build powerful AI-driven automation tools to optimize complex workflows.

Principles

Method

Use Claude Code with Apps Script to integrate AI into existing platforms like Gmail. Pull context from internal systems and public documentation via web search, then refine AI output through prompt iteration. Package solutions as plugins.

In practice

Topics

Best for: Automation Engineer, AI Engineer, Prompt Engineer

Related on AIssential

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Claude Blog.