Deekshita Senthil Kumar

Past Member

Deekshita Senthil Kumar

Junior Undergraduate Researcher I

deekshita@usf.edu

8/25/2025→1/9/2026


During my time in the lab, I contributed to the development of Bloom by working on text-to-speech integration, system configuration, and demo preparation. I evaluated multiple TTS models, including Piper and OpenAI GPT-4o, testing various inputs and measuring average latency to determine performance. Through this process, I found GPT-4o to be the strongest overall option, while Piper served as a reliable offline alternative.

I also configured Comitup to enable Bloom to create and manage its own Wi-Fi access point, allowing for seamless standalone connectivity during setup and deployment. This significantly improved usability by removing the need for an external network and supporting smoother user interaction. In addition, I helped produce the HRI 2026 demo video, which showcased Bloom’s real-time interaction capabilities, internal hardware structure, and overall system functionality in a realistic environment.

This experience strengthened my skills in system testing, technical evaluation, problem-solving, and working within a research-driven, collaborative environment. I’m incredibly grateful for the guidance and support from the professor, mentors, and lab members, and I’ll be carrying these experiences forward into my future academic and professional work. I highly recommend joining the lab to anyone looking to gain hands-on, real-world technical experience as it’s an amazing place to learn and grow!

— 1/13/2026

Hi! My name is Deekshita Senthil Kumar, and I am a senior majoring in Computer Science at the University of South Florida. My background combines data analysis, automation, and software development, highlighted by projects involving data quality tools, full-stack systems, and robotics using Python, PyTorch, SQL, and Flutter, bridging technical skills with research applications.

I am particularly interested in human-robot interaction, data-driven research, and how emerging technologies can be applied to create meaningful and intuitive user experiences. At RARE Lab, I am excited to contribute my skills in data analysis and software development to real-world research projects, while learning more about AI, robotics, and human-centered design.

I believe RARE Lab will be the perfect environment to refine my technical expertise, collaborate with a multidisciplinary team, and help advance research at the intersection of technology and human behavior.

— 8/24/2025


Publications

Xiangfei Kong, Rex Gonzalez, Nicolas Echeverria, Sofia Cobo Navas, Thuc Anh Nguyen, Divyamshu Shrestha, Sebastian Ramirez-Vallejo, Deekshita Senthil Kumar, Andrew Texeira, Thao Nguyen, Deep Akbari, Hong Wang, Hongdao Meng, Zhao Han

Bloom Preview: A Low-Cost LLM-Powered Social Robot

2026 ACM/IEEE International Conference on Human-Robot Interaction (HRI), Interactivity

,

2026

News

HRI 2026 LBR and Interactivity Acceptance

One LBR and one Interactivity paper were accepted to HRI 2026!

Congrats Hong & Bao on “Evaluating Dynamic Surface Compensation for Robots with Projected AR” (LBR) & Bloom team (Thuc Anh, Div, Sebastian, Hong et al.) on “Bloom Preview: A Low-Cost LLM-Powered Social Robot” (interactivity)!

Bellini College REU Symposium

Dr. Zhao Han was invited to talk about the benefits of REU from his extensive mentoring experience. Vidhi, Grant, Jasmine, Alissa, Deekshita, and Sebastian presented posters!