Primary AI development partner for Narratize's earliest generative features: document ingestion, technical briefs, research reports and scientific analysis.
Product innovation teams spent weeks turning research, internal documents and scattered ideas into technical briefs and decision-ready reports. Narratize needed generative features that experts would trust: grounded in their own sources, consistent in structure and fast enough to use every day.
Worked as the platform's primary AI development partner, turning the product vision into working AI capabilities for its earliest features.
Built document ingestion that brings research papers, internal files and notes into a structured knowledge layer.
Designed generation workflows for technical briefs, research reports and scientific analyses, combining OpenAI, Anthropic and Mistral models.
Connected research sources such as PubMed and arXiv, so drafts can draw on published work.
Added a voice storytelling agent: users speak, and hear a narrated, AI-generated story back, using Deepgram speech-to-text and text-to-speech.
Ran evaluation suites in CI on GitHub Actions, so every prompt or pipeline change was scored automatically. This was before LangSmith was mature enough to rely on.
Built the evaluation sets from curated examples of briefs and reports, and kept them in the repository alongside the code.
Supported releases across time zones, including late-night fixes when the client team needed them.
Document ingestion pipeline
Brief, report and analysis generation workflows
Multi-model orchestration
Research source connectors
Voice storytelling agent
CI evaluation suites
“I value your skill and dedication to your craft… I've worked with a lot of engineers over my career and have seen all different types, and I'm happy we found a group like yours which is of the highest quality.”
“[Our co-founder] 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. [He] helped translate our product vision into working AI capabilities, contributing to the development of some of Narratize's earliest product features.”
“I want to send a huge thank you to everyone who worked on making this feature possible, through all the troubleshooting and problem solving. It's so exciting to finally have this live and in front of users!”
Search technology, Digital marketing
A custom search and synthesis engine that combines several search APIs, live scraping and routed LLMs to deliver GEO insights, such as how a brand appears in AI answers, at a lower cost per query.
Market research, Consulting
An AI market-research platform with competitor analysis and a LangGraph-based TAM/SAM calculator, turning a founder's research process into self-serve tools.
Religious education, EdTech
A chatbot that answers questions about Islam from a curated library of religious texts, with signed-in users and conversation monitoring.
We can scope and build a system with comparable capabilities, shaped around your own workflows, data and integrations.
