Eindhoven University of Technology
I am currently in a doctoral program at the Eindhoven University of Technology (TU/e), The Netherlands. I am exploring the development and application of model-based engineering methodologies, in order to enable verification and validation early on in the product design-cycle for safety-critical systems. In particular, I am working on automating the composition of multiple simulators and mixed simulations-emulations to enable timing-analysis of the system.
Prior to this, I was employed as an SoC Architect at Qualcomm Inc., San Diego, CA. I was responsible for developing, integrating and supporting software simulation models of system-on-chip hardware components to construct "virtual platforms" that reduce the development cycle-time for cellphone chipsets, enabling and accelerating pre-silicon software development. After that, I joined NVIDIA (as a Senior Architect) where I was building a new "virtual platform development" team at the Bangalore Design Center back in India.
I am primarily interested in the domain of embedded system design, with a focus on system-level design methodologies and model-based design of system architectures.
My research interests include Embedded Systems Design, System-Level Design Methodologies, Hardware-Software Co-Design, System-on-Chip Architecture and Design, Multicore Architectures and Programming, Model-Based Design, Simulations & Virtual Prototyping, and Component-based Systems.
I graduated from Penn in 2012, with a MS degree in Embedded Systems.
During my time at EMBS, I received the opportunity to explore the realms of Computer Architecture, Electronic Design Automation, Real-time Embedded Systems, Cyber-Physical System Design, Computer Networks, Mechatronics, Robotics, as well as independent research with the faculty at PRECISE.
So far, what I have learned from my own experience working in the industry is that after students graduate and gain experience working in the real-world, they will realize that the methodologies employed to design such systems is quite often adhoc. Engineers gain individual skills from different courses, but a real-world product has to integrate all domains. What is not taught at most universities is the methodology of how to integrate all of these fields to build a quality product. Engineers in the real-world typically "hack" all these parts together, without a clear understanding of where all these domains fit.
I think this is why EMBS is so special. They place a strong emphasis on the methodology for integrating all of these fields. Classes like Rajeev's course (CIS 540) and Insup's course (CIS 541) provide the basis for realizing that logic can be implemented in all forms (hardware, software, firmware, etc.) and that these can be represented using mathematical notations. Once this is understood, students will learn to use the necessary tools to determine whether the system is designed to meet certain specifications. The process of doing this in the real-world is often primitive, relying solely only ordinary testing, without an emphasis on formal verification. This may be acceptable for products like cell-phones and PCs, but given that EMBS targets CPS that interact with humans, where errors can be fatal, ordinary testing is insufficient. It has been extremely beneficial in my career that I learned how to formally verify systems while I was at Penn.
Often time, my friends from other majors at Penn or other universities envy us. It is a great program because the faculty and staff at PRECISE really care about us, from advising us about our career path to the healthy food they order for us. Being able to walk into Rajeev’s or Insup’s office and they being very patient to listen and help us with our issues made a huge difference. Although I have graduated, whenever I email the staff and faculty at PRECISE, they always respond very promptly.