MNTN - AI Implementation Strategy

AI-augmented workflow system integrations for content strategy, article drafts and revisions, and content request weekly updates.

Built two AI workflow agents that accelerated article draft production by 48% and introduced the first ticket volume tracking KPIs across two content request dashboards.

Context

MNTN's support team runs an AI chatbot as the first line of response for help center queries. The chatbot's resolution rate is directly tied to article quality: missing coverage, outdated phrasing, or poor structure all translate into escalations to live support agents.

With 8 to 15 new content requests per week across six SME teams, maintaining quality at volume required more than good writing. It required building a system where AI-assisted workflows handled the repeatable parts of content production, including drafting, editing, SME question generation, and Slack reporting, so human editorial effort could focus on strategy, accuracy, and publication decisions.

I architected a modular AI agent system using Claude Desktop's Cowork environment, connecting directly to Slack, Google Drive, and the Intercom help center to automate each stage of the content lifecycle.

The workflow system is structured as a modular skill library. Each skill handles one stage of the content lifecycle and can be triggered independently or chained. This architecture keeps individual agents narrow and auditable, and makes the system easy to extend as content needs evolve.

My approach

  1. Map the content lifecycle

    • Identified every repeatable task in the content workflow, including request intake, status tracking, drafting, SME review, editing, and stakeholder reporting, then evaluated which stages were bottlenecks suited to AI augmentation versus which required human judgment.

  2. Build a modular skill architecture

    • Designed each AI agent as a standalone skill with a hardcoded output format, strict numbered steps, and explicit escalation paths for edge cases. Modular design meant skills could be used individually or chained, so a content request could flow from aggregation to drafting to reporting in a single session.

  3. Connect to live data sources

    • Integrated skills with live Slack Lists trackers, Google Drive style guide documents, and Intercom article records, giving each workflow access to current data rather than static snapshots. Resolved a root-cause data access issue where Slack Lists require CSV export because direct channel reads do not return list rows.

  4. Embed guardrails at the skill level

    • Wrote no-hallucination policies, SME escalation triggers, and source-tracing requirements directly into each skill's instructions, ensuring AI-generated content met editorial standards before any human review step, not after.

Collaboration

The AI workflow system was built alongside the existing cross-functional review process, not as a replacement for it. SME teams from Product, Engineering, and Customer Success remained the accuracy gate for every article. The workflows automated the parts they should not have to care about: status reads, draft generation, gap flagging, and reporting. That kept the SME relationship focused on substantive review rather than coordination overhead.

Results

  • Automated weekly status reporting replaced manual Slack updates. The system reads live Slack trackers, categorizes request health, and generates formatted summaries with priority flags and blockers every Wednesday and Friday.

  • AI-assisted drafting and editing using a custom content-writer AI agent with built-in skills that enforces MNTN's style guide, generates SME question sets for gaps, and produces Slack-ready edit summaries on approval, all within a single workflow run.

  • No-hallucination guardrails built into every workflow step. All AI-generated content must trace to source material, and unsourced claims surface as explicit SME questions rather than appearing as finished copy.

  • Transformed two manual content operations, accelerating draft production by 48% (across the PTV Help Center and QuickFrame Marketplace Help Center content) and building the first measurable KPI framework across two content request dashboards in Slack.