Guests
- Elliot Little, Product Manager, Zero Gravity
- Dan St. Paul, Software Engineer, Zero Gravity
What we cover in this episode
- Zero Gravity's mission: breaking down barriers to elite careers for disadvantaged UK students
- The "knowing-doing gap"—why students struggle to act even when they know what to do
- Why their first prototype (a job suitability summary) didn't create the "wow moment" they expected
- The decision to use text chat over voice input and why guided prompts beat empty text boxes
- Context management techniques: removing stale tool calls, summarizing history, exposing tools conditionally
- Using different models for different tasks (GPT-5 Nano for structured outputs, lighter models for quick replies)
- Safeguarding architecture: moderation endpoints plus external verification with Unitary
- Building a failure taxonomy through internal red team/green team exercises
- What's next: long-term memory management for multi-year student journeys
Links & References
Chapters
00:00 Introduction to Dan and Elliot
00:45 Zero Gravity's Mission and Impact
02:14 Introducing the AI Career Co-Pilot
04:01 Challenges Faced by Disadvantaged Students
06:49 Zero Gravity's Mentorship Program
09:14 Building the AI Career Co-Pilot
12:01 Early Prototypes and User Feedback
17:05 Refining the AI Career Co-Pilot
37:36 Introduction to Career Co-Pilot
38:02 Current Student Interactions
40:22 Technical Deep Dive
42:14 Context Management Challenges
44:43 Tool Call Optimization
51:48 Safeguarding and Moderation
57:52 Evaluating AI Performance
01:04:09 Future Directions for Career Co-Pilot
01:07:52 Concluding Thoughts
