I finally got it down to 497 words out of a word limit of 500. It used to be around 520.
Scientific exploration clearly excites you. Beyond our 3:1 student-to-faculty ratio and our intense focus on research opportunities, how do you believe Caltech will best fuel your intellectual curiosity and help you meet your goals?
The day I bought the desktop computer that currently hosts my website and a myriad of experiments, I slid my screwdriver down the warranty-void-if-broken seal and got to work swapping out different RAM and graphics cards, then inching my CPU to ever-faster clock rates. A few computer crashes later, I coaxed my computer to calculate pi to 32 million digits 2 seconds faster than it had previously before .
That was my naïveté four years ago, just a few weeks before I discovered that a team led by Prof. Harvey Newman has again broken Caltech's record for network data transfer rate, with a continuous data transfer speed of 339gb/s. However, this figure is backed not by simple tweaking of existing technology as I had done, but by major advances in signal processing, network architecture and data storage. After having read of this discovery, I immediately knew that I have to take my innate need for speed out of my basement and into Caltech.
I have a deep-seated curiosity in all computer systems from the most fundamental silicon chip, to digital signals, to the more abstract software. To perpetuate these interests, I aspire to join the MICS lab under Prof. Azita Emami. There, I can develop fundamental technologies like the Network-on-a-Chip paradigm to improve parallel processing or using light to encode signals with ever-faster bandwidth. Then, I can see my ideas physically manifested in silicon, waiting to be tested. Having two summers' experience in developing novel protocols of signal processing and digital communication with research in embedded systems, I am ready to fully utilize MICS to satisfy my desire to push the limits of computation.
Along with my designing of new hardware architectures, I beg to know how to best utilize them through software. That's why I aspire to take CS179-GPU Programming, in conjunction with my electrical engineering courses. Through learning how to exploit the unique capabilities of graphics processing units, I can write innovative algorithms to join the bigger effort to better and faster process big data. Meanwhile, learning about GPUs allows me to better understand the needs of the very applications that depend on hardware and thus, will supplement my microcontroller development by opening my mind to the many different computational systems.
I first-handedly know that the Caltech teaching style suits me very well as I have virtually taken EE156-Learning Systems in order to broaden my understanding of artificial intelligence. Through simply watching the lectures on Caltech's YouTube channel in my free time, I was able to glean ideas which I was able to better expand upon by reading online articles. My biggest test was when I applied what I learned in the form of a neural network-based letter recognition program on my TI84. When I glided my finger across the keypad of my calculator, "drawing" out the letter 'A', I was gratified to see that the correct letter popped onto the screen after 4 seconds. It was 1 second faster than my classical algorithm.
Thanks for any feedback! Out of a scale of 10, how would you rate this?
Scientific exploration clearly excites you. Beyond our 3:1 student-to-faculty ratio and our intense focus on research opportunities, how do you believe Caltech will best fuel your intellectual curiosity and help you meet your goals?
The day I bought the desktop computer that currently hosts my website and a myriad of experiments, I slid my screwdriver down the warranty-void-if-broken seal and got to work swapping out different RAM and graphics cards, then inching my CPU to ever-faster clock rates. A few computer crashes later, I coaxed my computer to calculate pi to 32 million digits 2 seconds faster than it had previously before .
That was my naïveté four years ago, just a few weeks before I discovered that a team led by Prof. Harvey Newman has again broken Caltech's record for network data transfer rate, with a continuous data transfer speed of 339gb/s. However, this figure is backed not by simple tweaking of existing technology as I had done, but by major advances in signal processing, network architecture and data storage. After having read of this discovery, I immediately knew that I have to take my innate need for speed out of my basement and into Caltech.
I have a deep-seated curiosity in all computer systems from the most fundamental silicon chip, to digital signals, to the more abstract software. To perpetuate these interests, I aspire to join the MICS lab under Prof. Azita Emami. There, I can develop fundamental technologies like the Network-on-a-Chip paradigm to improve parallel processing or using light to encode signals with ever-faster bandwidth. Then, I can see my ideas physically manifested in silicon, waiting to be tested. Having two summers' experience in developing novel protocols of signal processing and digital communication with research in embedded systems, I am ready to fully utilize MICS to satisfy my desire to push the limits of computation.
Along with my designing of new hardware architectures, I beg to know how to best utilize them through software. That's why I aspire to take CS179-GPU Programming, in conjunction with my electrical engineering courses. Through learning how to exploit the unique capabilities of graphics processing units, I can write innovative algorithms to join the bigger effort to better and faster process big data. Meanwhile, learning about GPUs allows me to better understand the needs of the very applications that depend on hardware and thus, will supplement my microcontroller development by opening my mind to the many different computational systems.
I first-handedly know that the Caltech teaching style suits me very well as I have virtually taken EE156-Learning Systems in order to broaden my understanding of artificial intelligence. Through simply watching the lectures on Caltech's YouTube channel in my free time, I was able to glean ideas which I was able to better expand upon by reading online articles. My biggest test was when I applied what I learned in the form of a neural network-based letter recognition program on my TI84. When I glided my finger across the keypad of my calculator, "drawing" out the letter 'A', I was gratified to see that the correct letter popped onto the screen after 4 seconds. It was 1 second faster than my classical algorithm.
Thanks for any feedback! Out of a scale of 10, how would you rate this?
