We are seeking a technically-minded builder to help create an early proof-of-concept AI knowledge system.
The goal is not to build a production software company initially. Instead, we want to rapidly test whether AI can help organize, search, analyze, and extract insights from a large library of educational, therapeutic, and training content.
This project sits at the intersection of AI, knowledge management, education, psychology, coaching, and human development.
We are looking for someone who enjoys building practical systems quickly, experimenting with AI tools, and translating complex ideas into working prototypes.
Initial Project:
Build a proof-of-concept system that can:
- Ingest video recordings.
- Generate transcripts.
- Organize transcripts into a searchable knowledge base.
- Identify recurring themes, patterns, and concepts.
- Support semantic search and question-answering across the content library.
- Surface insights and connections that may not be obvious through manual review.
- Generate summaries and thematic analyses.
- Allow retrieval of relevant information across multiple content sources.
Content Sources May Include:
- Educational videos.
- Teaching content.
- Retreat recordings.
- Therapy-related content.
- Books.
- Articles.
- Written educational materials.
- Training curriculum.
- Internal documentation.
- Podcasts and interviews.
Long-Term Vision:
The long-term vision is to create a secure, private AI knowledge system capable of learning from a curated body of content and supporting increasingly personalized interactions.
Rather than functioning as a generic chatbot, the system would ideally be able to:
- Understand themes and frameworks across a large body of content.
- Identify patterns and relationships between concepts.
- Retrieve relevant knowledge from multiple sources.
- Provide contextually relevant responses grounded in approved materials.
- Preserve and extend a particular educational, coaching, and therapeutic framework.
We are interested in exploring whether AI can help recreate aspects of an ongoing educational, mentoring, coaching, or therapeutic relationship through retrieval and synthesis of trusted source material.
This is not intended to replace human clinicians, teachers, or coaches. Rather, the goal is to explore how a carefully designed AI system can support continuity, reflection, learning, and access to knowledge.
Security & Privacy Requirements:
This project involves highly sensitive intellectual property and may eventually involve educational, coaching, and therapy-related content.
Security is a top priority.
Applicants should have experience or familiarity with:
- Private AI deployments.
- Secure cloud architecture.
- Encrypted storage.
- Role-based permissions.
- Data governance.
- HIPAA-aware design principles (preferred).
- Secure API integrations.
We strongly prefer approaches where:
- Uploaded content remains private.
- Proprietary content is not exposed publicly.
- Source material is not used to train public AI models.
- Ownership of all data remains with the company.
- Retrieval systems operate within a controlled environment.
- Access can be audited and managed internally.
One of the primary goals of this project is to create a closed knowledge environment that protects intellectual property while still benefiting from modern AI capabilities.
Preferred Skills:
- OpenAI API.
- Anthropic Claude API.
- Whisper or speech-to-text systems.
- Retrieval-Augmented Generation (RAG).
- Vector databases.
- Pinecone, Weaviate, Chroma, or similar tools.
- Python.
- Prompt engineering.
- AI workflow automation.
- Zapier, Make, or n8n.
- Notion integrations.
- Secure cloud infrastructure.
Nice to Have:
- Experience building AI knowledge bases.
- Experience building RAG systems.
- Experience with healthcare, education, coaching, or therapeutic content.
- Experience with private/self-hosted AI systems.
- Experience organizing large unstructured knowledge libraries.
- Experience working with sensitive or regulated data.
What Success Looks Like:
Within the first phase of the project, we would like to demonstrate that we can:
- Upload educational and video content.
- Generate accurate transcripts.
- Create a searchable knowledge repository.
- Identify themes and patterns across content.
- Retrieve meaningful responses grounded in source material.
- Maintain strong security and privacy controls.
Additional Notes:
We are far more interested in practical builders than theoretical experts.
The ideal candidate can quickly create a working prototype, communicate clearly, and balance technical capability with strong privacy and security practices.
