Summary
Divish Dayal is a Staff Machine Learning Engineer with 11 years of experience building production-scale ML systems that power retrieval, generative, recommendation, and classification solutions across advertising, sales enablement, and enterprise productivity. He has deep expertise in representation learning and transformer-based models, having designed production embedding pipelines and per-client model training workflows that serve real-time transcripts and large content repositories. At Highspot he helped deliver agentic AI that autonomously queries databases and executes APIs, and at Pencil he led multimodal creative AI for global brands, tying models directly to $1B+ in media spend. His work blends hands-on model training and fine-tuning with robust API, microservice, and cloud-native deployments to ensure scalable, observable ML in production. Notably, he moves between research and product: publishing early generative ad models and shipping enterprise-grade NLP pipelines that process millions of documents daily. Based in Vancouver, he combines a research-backed foundation from NUS and BITS Pilani with a practitioner’s focus on measurable business impact.
11 years of coding experience
7 years of employment as a software developer
BITS Pilani, Birla Institute of Technology and Science