Back to Work
Scaling Presales Research with AI
An end-to-end text classification and research pipeline that improved lead conversion and close rates.
The problem
AIREV needed to automate lead scoring and report generation at scale.
The challenge was processing long documents reliably and producing structured outputs teams could act on.
What I built / changed
- Built Revolution Engine: an end-to-end text classification service for training and deploying BERT models.
- Built PresalesAI: a GPT-4 powered research tool for structured analysis.
- Designed MapReduce-style LLM pipelines for long documents.
- Designed structured PostgreSQL schemas for data management.
Result
20% faster lead conversion and 35% higher sales close rates.
Stack / concepts
BERTGPT-4LangChainMapReduce
Interested in discussing engineering challenges?
Get in touch