[Google scholar] [DLBP] [LinkedIn]
I am super excited to join LG Electronics USA, Inc. as a Staff Security Research Engineer (Cryptography) on Jan 2023 – looking forward to working on amazing projects on lattice-based post-quantum cryptography, elliptic-curve cryptography, and machine learning for safety in vehicle-to-everything communications.
From Feb 2021 to Jan 2023, I worked as a Intel Noyce-Sponsored Postdoctoral Scholar at the Department of Computer Science, at the University of California, Davis. With three PIs, Profs. Chen-Nee Chuah, Zubair Shafiq, and Houman Homayoun, I co-lead the Intel Noyce-sponsored project on security and privacy aspects of ML/DL models on regular platforms as well as IoT/edge/embedded systems. Also, I have been leading projects on secure multi-party computation (MPC)-based privacy-preserving machine learning (PPML) for regular systems as well as the next-generation edge network systems that support in-network computing.
In Jan 2021, I have completed my Ph.D. from the Department of Computer Science at Indiana University-Bloomington under the supervision of Prof. Ryan Henry. We explored Private Information Retrieval (PIR), a cryptography-based Privacy-Enhancing Technology, to turn it from a theoretical construct to a useful tool in the privacy practitioners toolkit.
In 2018, I have pursued a research internship at the International Computer Science Institute, ICSI (affiliated with UC Berkeley). I worked in a Censorship Circumvention project with Dr. Sadia Afroz and Prof. Damon McCoy.
Before that, for May 2012 - Aug 2014, I worked as a Research & Development Engineer in the Offshore R&D Lab of Kona International, South Korea at Dhaka, Bangladesh. The company is specialized as a global provider of Smart Card Technology and Cryptographic Solutions.
Previously, I have earned my BS and Engg. degree in Computer Science and Engineering from Bangladesh University of Engineering and Technology (BUET) on April 1, 2012.
Gimme That Model!: A Trusted ML Model Trading Protocol.
A Bit More Than a Bit Is More Than a Bit Better: Faster (essentially) optimal-rate many-server PIR.
DNN Model Architecture Fingerprinting Attack on CPU-GPU Edge Devices.
Stealthy Inference Attack on {DNN} via Cache-based Side-Channel Attacks.
Querying for Queries: Indexes of Queries for Efficient and Expressive IT-PIR.
An efficient scanning based learning free algorithm for face detection.
Faster Optimal-rate Many-server Private Information Retrieval (PIR).
Protecting Tor Hidden Services with Tunicate Onion Descriptors.
Practical Evaluation of Proxy Distribution Mechanisms in Tor.
System of Knowledge: Private Information Retrieval.