Instagram story viewer> @umdcs> Posts

UMD Computer Science

@umdcs

1253

posts

5.5K

followers

117

following

Official feed for the University of Maryland's Department of Computer Science housed in the Brendan Iribe Center

POSTS STORIES REELS TAGGED

Download All

We call this the CS Walk 🎓🦾

Today, our @UMDCS undergraduate students cross the stage and step into what’s next as @univofmaryland graduates.

Congratulations, #CSTerps. You earned this moment. by @umdcs

274

4

2 months ago

🦾 And there you have it, they’re off to create the next big thing, advance research and change the world. @univofmaryland CS Terps🎓 by @umdcs

239

2

2 months ago

It’s an amazing feeling. 🎓

@univofmaryland CS grads make it look good. by @umdcs

87

1

2 months ago

According to the University Career Center’s annual graduation survey, @univofmaryland Computer Science graduates reported a 95% job placement rate for the class of 2025.

Among graduates with known outcomes, 78.5% entered the workforce and 16.2% continued their education. The median starting salary increased to $119,000, and more than 85% of respondents completed at least one internship during their time at Maryland.

Read more: Link in bio. by @umdcs

332

5

3 days ago

Three @univofmaryland computer science students received CMNS Alumni Network Endowed Undergraduate Awards to support their research, conference travel and career development.

Raghav Thind is attending the Summer@EPFL research fellowship in Switzerland, where he is continuing his development as a research scientist.

Nghia Truong used the funding to present his first-author paper at the 2026 IEEE International Conference on Acoustics, Speech, and Signal Processing in Barcelona. This summer, he is also studying transformer attention and helping build a multimedia quiz system designed to challenge artificial intelligence systems.

Priya Tyagi is continuing her work in the Spatial Biology Lab, where she is exploring how machine learning and imaging data could be used to predict tissue measurements, potentially reducing the time and cost of biological research.

Read more: Link in bio. by @umdcs

84

0

4 days ago

Online security and privacy depend on more than technology. They also depend on how people make decisions, share information and use digital tools.

@univofmaryland Professor of Computer Science Michelle Mazurek studies how people handle security and privacy in everyday systems. Her work looks at users, software developers and professionals responsible for protecting other people’s information.

In this Q&A, Mazurek discusses her path into computing and her research on ways to protect sensitive study data before it is shared with other researchers. She also explains the challenge of keeping data useful while making it harder to connect that information to a specific person.

Read more: Link in bio. by @umdcs

28

0

5 days ago

A secure AI model can still operate within an insecure system.

@univofmaryland researchers developed GRIEF, a testing tool designed to identify vulnerabilities in the software infrastructure that serves large language models to many users simultaneously.

When tested on two widely used open-source LLM serving systems, GRIEF uncovered 16 vulnerabilities, including cases involving leaked prompts, corrupted responses and denial-of-service failures.

The research was led by computer science doctoral student Yunze Zhao, advised by Professor Michelle Mazurek.

Read more: Link in bio. by @umdcs

13

0

9 days ago

🧬 A single biological sample can produce millions of DNA fragments, making it difficult for researchers to locate specific sequences within expanding genetic databases.

@univofmaryland computer science Ph.D. student Ryan Synk is developing LOCALE, a machine-learning system designed to search @nihgov's Sequence Read Archive. The system aims to identify related genetic sequences even when they contain mutations, biological variation or errors introduced during sequencing.

Advised by Professor of Computer Science Ramani Duraiswami, Synk is looking to reduce the system’s computational demands and test it across a larger portion of the federal archive.

Read more: Link in bio. by @umdcs

33

0

9 days ago

@univofmaryland CS alum Nakul Garg (Ph.D. ’25) received the 2026 ACM SIGMOBILE Dissertation Award for research on energy-efficient sensing technologies designed for devices with limited power, size and computing resources.

Advised by Associate Professor Nirupam Roy, Garg developed systems that use the physical properties of sound, radio and light to help small devices interpret their surroundings while reducing their reliance on large sensors, extensive computing resources and cloud services.

Garg is now an assistant professor of electrical and computer engineering at Rice University, where he continues to study ambient intelligence, mobile computing, embedded artificial intelligence and sensing systems.

Read more: Link in bio. by @umdcs

21

0

12 days ago

This summer, 27 undergraduates are applying computer science and mathematics to real-world research challenges through @univofmaryland’s Research Experience for Undergraduates in Combinatorics and Algorithms for Applied Research.

Selected from nearly 400 applicants, the students are working across seven research teams on projects involving evolutionary trees, artificial intelligence, parallel computing and Ramsey theory. The 10-week, @nsfgov-funded program pairs participants with UMD faculty and graduate student mentors while introducing them to collaborative, graduate-level research.

Read more: Link in bio. by @umdcs

193

3

12 days ago

The @univofmaryland Department of Computer Science remembers G.W. “Pete” Stewart, a faculty member from 1974 to 2008 whose work in numerical linear algebra and mathematical software helped shape scientific and engineering computation.

Stewart was one of the developers of LINPACK, a portable software package for dense matrix computations that became a standard tool for solving systems of linear equations and computing matrix decompositions. Its benchmark codes were also used for years to evaluate computer performance.

Read more: Link in bio. by @umdcs

15

0

16 days ago

As AI systems process larger amounts of information, video remains one of the toughest challenges.

@univofmaryland computer science Ph.D. student Manan Suri is working to change that through Frames2LoRA, a project that explores zero-token video understanding for more efficient video summarization, question answering and real-time decision-making.

Advised by Distinguished University Professor Dinesh Manocha, Suri’s research aims to reduce answer-time visual-token load, lower latency and make repeated questions about the same video more efficient.

Read more: Link in bio. by @umdcs

41

0

17 days ago