Saumya P.

Monitoring and Observability (M&O): Responsible for Meta’s service reliability, the Detection Foundation team develops and manages core detection products, monitoring millions of detectors to enable rapid issue detection across key platforms like AI, Developer Infrastructure, Instagram, Facebook, and WhatsApp.

Led the project within the Intelligence and Knowledge Discovery Lab under Prof. Xiang Ren and Shushan, expanded HuggingFace’s BigCode framework via in-context learning to enable transfer of models from high-resource coding languages to low-resource ones.

Engineered a customer churn prediction framework with data mining techniques for interpretability, examining 10+ pivotal factors. Simultaneously, executed TensorFlow-powered Deep Reinforcement Learning (DDPG and DDQN) to craft an adept stock trading agent across two separate initiatives.

Selected among the top 300 students in India to attend Amazon ML Summer School 2022 - an integrated learning program on key Machine Learning topics by Amazon ML Scientists.

• Got selected as a Women Techmakers Engineering Fellow, formerly known as WE TalentSprint Scholar, among the top 500 participants consisting of over 27,000+ applicants across India.