How to create your own AI performance coach: Optimizing your nutrition, recovery & injury management

Lucas Werthein, the COO and co-founder of Cactus, shares how he built a personalized AI wellness coach using ChatGPT to optimize his athletic performance while managing past injuries. After multiple surgeries on his knees, shoulder, and foot, Lucas created a system that synthesizes data from medical imaging, blood tests, wearable devices, and nutrition plans to provide personalized recommendations. His AI coach helps him balance competitive tennis, weightlifting, and running a company while maintaining his goal of “feeling 25 in a 40-year-old body.” Lucas demonstrates how this approach transforms siloed health information into actionable insights that protect joints, optimize recovery, and extend peak performance. *What you’ll learn:* 1. How to configure a ChatGPT with multiple data types, including MRIs, x-rays, blood tests, and wearable metrics, to create a comprehensive health profile 2. A framework for setting clear performance boundaries that prioritize joint protection, energy optimization, and injury prevention 3. Techniques for using AI to balance nutrition around special events like social dinners while maintaining performance goals 4. How to use images and videos to get AI feedback on physical symptoms and injury recovery timelines 5. A method for validating and contextualizing medical advice by having AI synthesize information from multiple health-care providers 6. Why creating clear rules and anti-prompts helps AI deliver practical, evidence-based recommendations instead of trendy supplements or extreme protocols *Copy Lucas’s Health Coach Prompt:* https://www.lennysnewsletter.com/p/how-to-create-your-own-ai-performance-coach *Brought to you by:* WorkOS—Make your app enterprise-ready today: https://workos.com?utm_source=lennys_howiai&utm_medium=podcast&utm_campaign=q22025 Google Gemini—Your everyday AI assistant: https://ai.dev/ *Where to find Lucas Werthein:* Website: https://cactus.is/ *Where to find Claire Vo:* ChatPRD: https://www.chatprd.ai/ Website: https://clairevo.com/ LinkedIn: https://www.linkedin.com/in/clairevo/ X: https://x.com/clairevo *In this episode, we cover:* (00:00) Introduction to Lucas’s athletic background and injury history (04:55) The challenge of synthesizing siloed health data (06:11) Building a GPT to optimize performance and recovery (09:57) Demonstrating the data types integrated into the AI coach (13:54) Configuring the GPT with clear performance goals and boundaries (16:31) Setting realistic expectations for the AI coach (17:50) Creating nutrition, training, and recovery frameworks (21:47) Establishing hard boundaries and anti-prompts (24:25) Example: Managing nutrition around special events (27:30) Accessibility and affordability of on-demand coaching (28:24) Practical examples and real-life scenarios (29:31) Using AI for injury management and recovery planning (34:19) Validating expert opinions and translating medical advice (37:25) Vision for the future of AI in personal health coaching (43:27) Other AI workflows: synthetic clients and AI co-founders (48:48) Final thoughts on AI reliability and evolution *Tool referenced:* • ChatGPT: https://chat.openai.com/ *Other references:* • InBody scan: https://inbodyusa.com/ • Whoop: https://www.whoop.com/ _Production and marketing by https://penname.co/._ _For inquiries about sponsoring the podcast, email jordan@penname.co._

Lucas WertheinguestClaire Vohost

Nov 23, 202551mWatch on YouTube ↗

WHAT IT’S REALLY ABOUT

Build a personal AI coach by unifying your health data

Lucas shares how repeated sports injuries and the overwhelm of siloed health data (wearables, labs, imaging, PT guidance, nutrition plans) pushed him to build a custom “WellCoach GPT.”
He shows how he manually feeds diverse inputs—Whoop CSVs, sleep/strain, journals, workouts, InBody scans, bloodwork, plus MRI/X-ray imaging—so the model can produce personalized, high-ROI recommendations.
A major emphasis is prompt design: clear goals (pain-free performance, joint safeguarding), realistic expectations, and strict “anti-prompts” that prevent risky or unproven advice.
The conversation expands into a future vision where AI becomes the missing synthesizer in healthcare—helping patients and doctors communicate better via continuously captured data—while also touching on related workflows like synthetic clients and an AI co-founder for brainstorming.

IDEAS WORTH REMEMBERING

5 ideas

The core problem isn’t missing health data—it’s missing synthesis.

Lucas already had experts and measurements (PT, doctors, labs, Whoop, scans), but struggled to unify them into coherent daily actions. His GPT acts as a “performance strategist” that consolidates inputs into prioritized next steps.

Multimodal context (including imaging) changes the quality of coaching.

Beyond workouts and food logs, adding MRI/X-ray files gives “under-the-hood” structural context. This supports more conservative, joint-protective recommendations aligned with injury history.

A high-performing AI coach starts with realistic goals and constraints.

Instead of “make me elite,” Lucas defines outcomes like pain-free movement, joint safeguarding, and sustainable output while running a company. That clarity helps the model recommend what’s feasible and relevant this week.

Frameworks beat tips: define nutrition, training-load, recovery, and feedback loops.

He encodes principles like stable glucose/low inflammation, load management using readiness/HRV, and sleep-first recovery. The AI then adapts tactics without reinventing the philosophy every chat.

Hard boundaries (anti-prompts) reduce risk and ‘biohacking drift.’

Lucas explicitly forbids pushing intensity on low-recovery days, novelty therapies, and unproven supplements, and instructs the coach to act on red flags. This keeps advice accessible, evidence-oriented, and safety-first.

WORDS WORTH SAVING

5 quotes

The problem isn’t the lack of data, the problem is the lack of synthesis and putting it all together.

— Lucas Werthein

I’m wanting to demand of my body to feel like twenty-five in a forty-year-old’s body.

— Lucas Werthein

Not only is this how I’m performing on an output basis, but this is… the structural setup under the hood.

— Claire Vo

No fluff, no hacks, nothing that hasn’t been proven.

— Lucas Werthein

In five years or less, everyone will have access to a personal AI health coach… not to replace doctors, but to help us show up more informed.

— Lucas Werthein

Health data silos and lack of synthesisCustom GPT setup with multimodal inputs (labs, wearables, imaging)Prompt architecture: role, goals, frameworks, toneHard boundaries and anti-prompts (safety, evidence, ROI)Nutrition planning around real-life eventsInjury management: validating and translating expert guidanceFuture of AI in healthcare and continuous passive sensingAdjacent workflows: synthetic clients, AI co-founder

High quality AI-generated summary created from speaker-labeled transcript.

Get more out of YouTube videos.

High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.