Atchuta Ram Aravilli

SWE @ Visa Inc

prof_pic.jpg

Bangalore, India

I am currently a Software Engineer at Visa in Bangalore, where I work on building resilient financial systems and applying machine learning to real-world challenges. My broader research interests lie in Deep Learning and Trustworthy AI, with a specific focus on AI Safety, explainability, and privacy-preserving machine learning.

Prior to my current role, I was an undergraduate research intern at the Institute for Development and Research in Banking Technology (IDRBT), where I was advised by Prof. Ravi Vadlamani. During that time, I worked on deep reinforcement learning for financial forecasting and graph-based anomaly detection.

For a closer look at my research work, please visit my publications page.

news

Nov 12, 2025 Paper on Enhancing Safety in Agentic VLMs accepted at NEURIPS Workshop 2025!
Nov 08, 2025 Paper on Preserving Privacy in LLMs accepted at NEURIPS Workshop 2025!

selected publications

  1. JIKM
    DRL_ABS.png
    Deep reinforcement learning for financial forecasting in static and streaming cases
    Aravilli Atchuta Ram, Sandarbh Yadav, Yelleti Vivek, and Vadlamani Ravi
    Journal of Information & Knowledge Management, 2024
  2. NeurIPS
    RES_FM.png
    Constrained Decoding for Privacy-Preserving LLM Inference
    Aravilli Atchuta Ram
    In Socially Responsible and Trustworthy Foundation Models at NeurIPS, 2025
  3. NeurIPS
    VLM_ABS.png
    From Vision to Action: Enabling Real-World Agentic VLMs
    Aravilli Atchuta Ram
    In 1st Workshop on VLM4RWD at NeurIPS, 2025