I am a second-year Ph.D. student in Computer Science and Engineering at Indian Institute of Technology, Bombay, advised by Prof. Supratik Chakraborty and Prof. S Akshay.
Prior to joining IIT-B, I was working as a research assistant (2022-24) guided by Prof. Yizhou Sun at University of California, Los Angeles during and after having completed my Master's in Computer Science (2021-23).
I also completed a stint of 3 years (2017-20) as a Software Engineer at Samsung Research Institute, Noida soon after I graduated in Electronics engineering from Indian Institute of Technology (Banaras Hindu University), Varanasi (2013-17).
Automated Functional Synthesis is the problem of synthesizing outputs as functions of inputs, while satisfying a declarative relational specification between inputs and outputs. In other words, provided synthesis of such a system is complete, it is guaranteed to meet the high-level specifications it is designed for, and hence correct-by-construction.
Recent advances in this problem has served as a meeting place of techniques rich in knowledge compilation, automated reasoning and learning.
While the black-box AI systems have gained upsurge, it is important to realize that they are far from being correct and increasingly even more so in cases when they run into safety critical pipelines.
Complexity-theoretic arguments indicate that it is extremely unlikely that the problem of synthesis has any polynomial-time algorithm. Our work focuses on constructing systems that can achieve correctness-by-construction. What makes these algorithms tick? What are the open problems? See this!
Synthesis
Formal Methods
Knowledge Compilation
Automated Reasoning
Neurosymbolic AI
I had opportunity of serving as a Teaching Assistant for the following courses:
I also had a brief yet fulfilling opportunity of serving as a Research Assistant under the guidance of:
Prof. Rafail Ostrovsky (2022-23),
Prof. Akihiro Nishi (2022-23),
Prof. Tao Gao (2023-24),
Prof. Yizhou Sun (2023-24).
