Sarmishta Velury
Source-listed: Engineer · Google DeepMind
Record snapshot: Sep 3, 2026. Coverage and affiliations may be incomplete or historical.
Selected work
4- Gemini: A Family of Highly Capable Multimodal ModelsSaved credit: Contributor · 2023Explore people connected to this work →
- Scalable Reinforcement Post-Training Beyond Static Human Prompts: Evolving Alignment via Asymmetric Self-PlaySaved credit: Contributor · 2024Explore people connected to this work →
- Reward-Guided Prompt Evolving in Reinforcement Learning for LLMsSaved credit: Contributor · 2025Explore people connected to this work →
- Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic CapabilitiesSaved credit: Contributor · 2025Explore people connected to this work →