Skip to content
ClaudeClaude

How Anthropic uses Claude in GTM Engineering

Sales reps drown in administrative work, from digging through scattered documentation to answering customer emails late into the night. Jared Sires, GTM Product Manager, shares how he went from account executive with zero coding experience to building CLAFTS (Claude Drafts), an AI email assistant powered by the Claude API that drafts customer replies in his own voice, with Claude Code. Check out the full case study to learn more: www.claude.com/blog/how-anthropic-uses-claude-gtm-engineering

Jared Siresguest
Jun 5, 20262mWatch on YouTube ↗

CHAPTERS

  1. 0:00 – 0:30

    From account executive overload to rethinking workflows with Claude

    Jared describes the surprise of becoming a GTM architect and contrasts it with the reality of being a startup AE managing hundreds of accounts. He frames the problem as time lost to email, Slack, and document searching—setting up why Claude became central to his workflow redesign.

    • Unexpected transition toward a more technical GTM role
    • AE life: managing ~600–700 accounts at peak
    • Long nights spent replying to emails and searching for info
    • Pain caused by scattered knowledge across Slack/Docs/Drives
  2. 0:30 – 0:46

    “How could I bring Claude into the loop?”—the origin of CLAFTS

    He pinpoints the turning point: deciding to integrate Claude into daily customer communications. This leads to building a tool focused on generating better, faster customer emails.

    • Identifies repetitive email work as the primary bottleneck
    • Decides to integrate Claude directly into the communication loop
    • Begins building internal solutions around email generation
  3. 0:46 – 1:03

    Introducing CLAFTS (Claude Drafts): inbox management with prompts + patterns

    Jared explains what CLAFTS is and what it does: an inbox-drafting system that combines context, a system prompt, and pattern matching. He positions it as a repeatable workflow any AE could follow.

    • CLAFTS = Claude Drafts
    • Goal: manage an inbox with AI-assisted drafting
    • Uses context knowledge + system prompt + pattern matching
    • Designed around a typical AE workflow
  4. 1:03 – 1:18

    Workflow step 1: generating a role-based system prompt

    He demonstrates the first part of the CLAFTS workflow: building a system prompt tailored to the user. The prompt captures identity, role, and what the AE helps customers with, creating consistent behavior across drafts.

    • Start by creating a system prompt
    • Inputs: name, role (e.g., Account Executive), customer value proposition
    • System prompt generation standardizes responses
    • Prompting happens via the Claude API behind the scenes
  5. 1:18 – 1:33

    Workflow step 2: context retrieval via Google Docs and web sources

    Next, Jared shows how CLAFTS retrieves relevant context automatically, using integrations like Google Docs and potentially web URLs. This replaces manual searching through internal sources and documentation.

    • Context retrieval uses a Google Docs integration
    • Can also fetch context from web URLs/Claude Docs sites
    • Adds a doc as the context source for the draft
    • Automates knowledge gathering that used to be manual
  6. 1:33 – 1:43

    Generating an email draft with internal + external knowledge automatically loaded

    With context connected, Jared generates an email draft and explains how Claude pulls from the website, documentation, and internal context. The key value is that the necessary information is loaded up front so drafting can start immediately.

    • Select an email and click “generate draft”
    • Claude retrieves info from company website and documentation
    • Also uses internal context previously added
    • Relevant information is loaded at the top to guide drafting
  7. 1:43 – 1:54

    Business impact: faster responses and more technical customer conversations

    He highlights the qualitative improvement: he can speak more technically with customers without hunting through Slack or docs. The system streamlines response quality and speed simultaneously.

    • Enables more technical, confident customer communication
    • Eliminates manual searching across Slack and Google Docs
    • Improves response consistency by leveraging centralized context
    • Makes customer support and sales communication more efficient
  8. 1:54 – 2:04

    Time savings and scale: hours per day saved and thousands of emails sent

    Jared quantifies the results: CLAFTS saves him multiple hours daily and has been used to send thousands of emails. This demonstrates both personal productivity gains and organizational scalability.

    • Saves ~2–3 hours per day on email responses
    • System has generated/sent thousands of emails
    • Automates repetitive drafting while maintaining quality
    • Shows measurable GTM productivity lift
  9. 2:04 – 2:23

    Career transformation: lowering the technical barrier to become a GTM architect

    He closes by explaining how Claude lowered the technical barrier, enabling him to design products and partner with engineers to finish implementation. That capability shift led directly to his new role and a changed career trajectory.

    • Technical barrier dissolves with Claude-assisted building
    • Can prototype/design solutions and then hand off to senior engineers
    • Transitioned into a go-to-market architect role
    • AI tooling reshaped his career path and scope of impact

Get more out of YouTube videos.

High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.