Thanks for the nice words! The technical work is split in three pieces: 1) LLM prompt and chaining for script generation 2) workarounds for some TTs bugs (not

Thanks for the nice words! The technical work is split in three pieces: 1) LLM prompt and chaining for script generation 2) workarounds for some TTs bugs (not that many, elevenlabs is amazing) 3) Stitching all moving pieces together Nothing advanced for tech, but requires a lot of experimentation as neither LLMs nor TTS are deterministic which create headaches for prod.