About the Role We are building a digital avatar - a highly personalized, interactive AI representation that captures his voice, persona, and knowledge. You will be responsible for implementing and configuring a SaaS-based avatar platform to bring this persona to life. The avatar will interact naturally with users, provide guidance, and showcase Mark’s unique personality while querying a structured knowledge base built from his books, videos, and other intellectual property. This role is SaaS-first but must be executed in a way that allows future migration to an open-source AI infrastructure without redoing the knowledge base or persona setup. Key Responsibilities Configure and deploy Mark’s avatar using SaaS AI tools for voice, video, and LLM-based conversational capabilities. Integrate Mark’s persona, tone, and style consistently across interactions. Connect the avatar to a structured knowledge base (books, video transcripts, writings) for dynamic querying. Maintain the knowledge base in a platform-agnostic format, ensuring portability for potential open-source deployment. Implement efficient retrieval of relevant context to feed into the SaaS LLM while maintaining performance and responsiveness. Collaborate with the team to ensure long-term maintainability, including logging, monitoring, and iterative persona updates. If experienced: provide guidance on hybrid architectures, memory persistence, or open-source migration strategies. Required Skills Proven experience with SaaS AI platforms (e.g., HeyGen, ElevenLabs, OpenAI API, Anthropic, etc.). Strong LLM integration skills and experience with prompt engineering. Familiarity with knowledge base integration, including vector databases (Chroma, Pinecone, Weaviate, etc.) or similar queryable systems. Ability to structure data and workflows for future portability outside the SaaS platform. Experience with media management: images, video clips, and voice assets. Comfortable collaborating with teams who are not AI engineers and explaining technical choices clearly. Preferred / Nice-to-Have Skills Familiarity with open-source LLM deployment (LLaMA, Falcon, Mistral, etc.). Experience designing hybrid AI pipelines or multi-layer AI architectures. Knowledge of scalable query optimization, caching, and API orchestration. Why This Role is Unique Work on a high-profile, personality-driven AI avatar from day one. Collaborate on a SaaS-first solution with future-ready portability to open-source. Opportunity to experiment with intelligent retrieval and knowledge integration in a real-world deployment. How to Apply Include examples of past AI avatars, LLM integrations, or SaaS AI work. Provide a short outline of how you would structure avatar's knowledge base for portability. Optional: Include links to any open-source contributions, hybrid architectures, or projects demonstrating familiarity with LLM deployment and persona modeling.