Vikas Agrawal

Vikas Agrawal

Consulting Data Scientist, AI and Applied Science

Oracle

Vikas Agrawal is a senior principal data scientist at Oracle Analytics Cloud. His role is to design, develop, and deploy AI applications in ERP, SCM, HCM, CX, and MFG across Fusion and NetSuite customers. These models need to get automatically generated, selected, and deployed for each customer as their data becomes available, and the models need to keep themselves updated as the world drifts. His research includes domain knowledge-driven, intelligent agents that compose available tools to solve complex problems, with approximate seeding of heuristics, hypotheses, relationships, and task hierarchies through critiqued LLMs and LMMs. Vikas is an electrical engineer turned computer scientist from IIT Delhi who has worked at CalTech, Intel Corporation, and Infosys. Currently, Vikas helps solve second- and third-year Computer Science problems for his son and daughter.

Deep learning AI models have mastered unstructured data problems including vision, speech and language, with foundation models available. Yet structured enterprise data—the bread and butter of actual business operations—remains largely untapped by foundation models. We discuss the new frontier of Relational Deep Learning which lets you skip months of custom ML development, and instead issue predictive queries directly over your relational databases—no schema flattening, no feature engineering, no per-task models required. The queries lead to generated AI predictions.

In this technically rich session, we’ll unpack:

  • How relational data is turned into temporal heterogeneous graphs.
  • The role of Relational Graph Transformers in learning across complex schemas.
  • The mechanics of in-context learning for structured data.
  • PQL (Predictive Query Language): a declarative interface for classification, regression, and recommendation—right inside your database.
  • Benchmarks demonstrating how this approach outperforms ML pipelines—zero-shot.
Read More

Managing and scaling ML workloads have never been a bigger challenge in the past. Data scientists are looking for collaboration, building, training, and re-iterating thousands of AI experiments. On the flip side ML engineers are looking for distributed training, artifact management, and automated deployment for high performance

Read More

Managing and scaling ML workloads have never been a bigger challenge in the past. Data scientists are looking for collaboration, building, training, and re-iterating thousands of AI experiments. On the flip side ML engineers are looking for distributed training, artifact management, and automated deployment for high performance

Read More
Book tickets View all speakers

We use cookies essential for this site to function well. Please click to help us improve its usefulness with additional cookies. Learn about our use of cookies in our Privacy Policy & Cookies Policy.

Show details