How it's delivered
Dusk Atlas operates through a hybrid human-platform model: trained undergraduate and graduate students from UT Austin and St. Edward's University serve as Digital Memory Guides, paired one-to-one or in small cohorts with Austin seniors aged 65+. Guides conduct weekly in-home or neighborhood-based sessions — visiting senior living communities, libraries, and community centers across Austin ZIP codes — where they facilitate storytelling, assist with recording oral histories, and scaffold digital publishing skills. A small full-time coordination staff (2-3 people) manages guide training, senior intake, session scheduling, and archive curation. Community partners including Austin History Center, Austin Public Library, and local neighborhood associations provide venue support and archival legitimacy.
How it scales
Growth follows a neighborhood-by-neighborhood expansion model, beginning in high-density senior areas such as 78745, 78723, and 78756. Each new neighborhood cohort is seeded by recruiting 4-6 student guides through UT and ACC service-learning pipelines, allowing low-cost labor scaling tied to academic calendars. Revenue diversification across neighborhood association grants, ACC/UT partnership stipends, Austin city cultural funding, and optional family subscription tiers enables incremental geographic expansion without venture dependency. A self-reinforcing archive — as published senior stories attract community attention — drives organic senior and guide recruitment.
Key bottleneck
The critical operational bottleneck is guide retention and quality consistency. Student guides cycle out each semester, creating recurring onboarding load and relationship disruption for seniors who depend on continuity and trust. Each guide departure risks abandoning a senior mid-story, undermining the program's emotional core. This requires a structured guide transition protocol and a minimum 2-semester commitment policy to be enforced rigorously.
Platform
Dusk Atlas depends on a lightweight custom web platform — built on WordPress or a similar low-code CMS — hosting the public-facing neighborhood story archive, with session scheduling managed through a tool such as Calendly or a purpose-built coordinator dashboard. Audio and video oral history files are stored on archival-grade cloud infrastructure. Secondary dependency exists on third-party recording tools (smartphone apps, Otter.ai for transcription) used during field sessions. Platform risk is low given commodity infrastructure, but the archive itself becomes the irreplaceable proprietary asset over time.