questions & curiosities

• What does dynamic "proof of work" or "demonstration of skills" look like in the age of AI?
  • The half-life of skills will likely decrease with new tech, but current ways of demonstrating skills (for learning or recruiting purposes) are static.
  • Personalization layer - Amazon has a really good idea of who I am as a consumer (loves books and sweets), and Netflix knows that I like lighthearted watches. Why can't we have a shared data layer for learning and recruiting? Why are they silo'ed?
  • In 6 weeks, I visited 10 campuses to meet student builders and founders. There's a greater appetite to explore via gap semesters - which I love. A big question - as people go in and out of school/work, how can we capture learnings and context of individuals?
• How might human learning inform embodied intelligence?
• What enablement layers are needed for accurate skills inventories and somewhat related, proliferation of physical intelligence (underwriting risk, human-robot interactions)

smaller questions (for now)

• How is taste formed?
• Advertising models are based on human clicks and views. What happens with more agents?
• Learning more about chip design, robotics software, brain-computer interface

drafts & tinkering

• Various interviews on talent matching and lifelong learning
  • Thank you to Harvard Graduate School of Education's Entrepreneurship Fellowship for funding!
• Reimagining higher education - writing in progress
  • Challenging notion that higher educations should be 2-4 years, and should really support learning in and outside of traditional classrooms. Largely inspired by Stanford Design School's 2015 exploration.
• Design. Designing thank you cards from scratch
• How do you capture "learning everywhere" - rough demo here
  • I learn so much in conversations and liminal moments - but it's not easily captured.
  • The product intent is that users can email, text, or manually input interesting conversations and learnings, to be aggregated in one place. As users compile different learnings, they will be able to see common themes and interests, and be recommended with specific skills and additional resources.
  • Learnings & limitations - high user input and onboarding required, data may not be robust enough for thoughtful recommendations