Co-founder project · Research and innovation
Narratize
Document ingestion and drafting for technical briefs and research reports.
- Built by
- Muhammad Zubair, Sortup co-founder
- Sector
- Research and innovation
- What was built
- AI
- Stack
- OpenAI
- Anthropic
- Mistral
- Django
- Streamlit
- Deepgram
- Client
- An innovation intelligence platform.
- Engagement
- Zubair built generative AI features for ingesting documents and drafting technical briefs, research reports and scientific analyses, plus a voice storytelling agent.
Proof you can open
The challenge
Innovation teams spend a lot of time turning research and scattered documents into decision-ready briefs and reports.
The solution
Ingestion and drafting workflows on OpenAI, Anthropic and Mistral models with evaluations, built on Django and Streamlit, and a voice agent using Deepgram speech-to-text and text-to-speech.
The impact
- Briefs and reports drafted from ingested documents.
- Several model providers behind one workflow.
- A voice agent that narrates stories from speech.
He was a thoughtful and collaborative partner who took the time to understand what we were trying to build and the problems we wanted to solve. Zubair helped translate our product vision into working AI capabilities, contributing to the development of some of Narratize’s earliest product features.
Building something like Narratize?
No pitch deck. An engineer, and a straight answer.