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

• Zero Gravity
• Unitary – AI-powered content moderation
• Blue Dot Impact AI Safety Course – free AI safety course Elliot recommended

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