Haarika Manda

I am a fourth-year PhD student in Computer Science at UC Santa Barbara, co-advised by Elizabeth Belding and Arpit Gupta.

I’m interested in how machine learning can help us understand the Internet, and how working with real network data can help us build better models. My research focuses on representation learning for network telemetry, bringing together my interests in machine learning, Internet measurement, and networking systems.

In my work on TurboTest, I use a dual-stage time-series model to decide when a speed test has collected enough information to stop. I also study network foundation models through netFound. My ongoing work on PRISM explores how well learned representations carry across tasks and network conditions.

I’m an NSF Graduate Research Fellow. My earlier SIGCOMM ’24 work on broadband deployment received the 2025 IRTF Applied Networking Research Prize. I also received the UCSB Department of Computer Science’s Outstanding Publication of the Year award in June 2025.

hmanda@cs.ucsb.edu CV Google Scholar GitHub LinkedIn

Selected publications

All publications

TurboTest: Learning When Less is Enough through Early Termination of Internet Speed Tests

Haarika Manda, Manshi Sagar, Yogesh, Kartikay Singh, Cindy Zhao, Tarun Mangla, Phillipa Gill, Elizabeth Belding, Arpit Gupta

USENIX NSDI, 2026

A dual-stage time-series model learns when a speed test has collected enough evidence to stop, balancing measurement accuracy and data cost.

Evaluated on 1 million M-Lab tests · 1.8–4.4× higher data savings than a BBR-signal baseline, with lower median error

Paper Talk & details

The Efficacy of the Connect America Fund in Addressing US Internet Access Inequities

Haarika Manda*, Varshika Srinivasavaradhan*, Laasya Koduru, Kevin Zhang, Xuanhe Zhou, Udit Paul, Elizabeth Belding, Arpit Gupta, Tejas N. Narechania

ACM SIGCOMM, 2024

Assessing ISP deployment and service delivery under the Connect America Fund.

* Equal contribution · 2025 IRTF Applied Networking Research Prize

Paper ACM Project

netFound: Principled Design for Network Foundation Models

Sylee Beltiukov, Satyandra Guthula, Haarika Manda, Jaber Daneshamooz, Wenbo Guo, Walter Willinger, Arpit Gupta, Inder Monga

arXiv preprint, 2026

A network foundation model designed to learn reusable representations of traffic.

Paper arXiv

Pramana: A Composable, Domain-Specific Backend for Empirical Networking Research

Jaber Daneshamooz, Eugene Vuong, Alagappan Ramanathan, Manni Moghimi, Haarika Manda, Satyam Kumar, Snithik Thode, Satyandra Guthula, Sylee Beltiukov, Dongsu Han, Tarun Mangla, Sangeetha Abdu Jyothi, Walter Willinger, Arpit Gupta

ACM HotNets, 2026 · Accepted

Specify network conditions and application workflows to generate telemetry for repeatable experiments.

Paper arXiv

Rethinking Network Foundation Models: From Bigger Models to Better Data

Jaber Daneshamooz, Haarika Manda, Manni Moghimi, Sylee Beltiukov, Walter Willinger, Arpit Gupta

NTC Reloaded · CoNEXT workshop, 2026 · Accepted

Manuscript not yet public.

Workshop program

Bridging multimodal microscopy for advanced characterization on nuclear fuel using machine learning

Haarika Manda*, Liang Zhao*, Rahul Reddy Kancharla, Xianghui Xiao, Charith Purushotham, Yalei Tang, Peng Xu, Tiankai Yao, Fei Xu

Frontiers in Mechanical Engineering 11, 1619834, 2025

* Equal contribution.

Paper

Measuring Broadband Policy Success

Haarika Manda, Varshika Srinivasavaradhan, Laasya Koduru, Kevin Zhang, Xuanhe Zhou, Udit Paul, Elizabeth Belding, Arpit Gupta, Tejas N. Narechania

Harvard Law Review Blog, July 16, 2024

Paper