Internship Experience at Texas Instruments

I still remember the first time I walked through those doors at Texas Instruments. Long rows of cubicles ran off both sides as I stood there observing others at work. I saw oscilloscopes, logic analyzers, FPGAs and PCBs, soldering irons, transistors scrambled on the desks with people intriguingly looking up at a distorted sine wave on the screen or writing up some code to implement their logic. And I thought “What more an electrical engineer can ask for? We have got all the hardware in the world, the best minds at work to guide us, all the resources to create something new, something helpful to society.”

That’s when I woke up from the dream, it was 7:49 am. Yes, it was all a dream. The world has been struggling for more than a year because of COVID 19. So, no more office tours and meeting rooms, only virtual orientations and WebEx meetings. Coming to what really happened.

I was to work on an interface called JESD204C (A technical side note, JESD204C is a high-speed serial interface between ADC’s and FPGA devices. It can sustain lane rates up to 32Gbps, so calling it fast is an understatement.) My first task was to understand a project on JESD204C that was written previously by members of the Texas Instruments Communication Lab. It was my first time seeing such a humungous project with multiple modules full of code. My mentor Mr. Avinash taught me how to steer through this jungle of code and build a map along the way. By a map, I mean a data path of how the data gets processed in FPGA to finally retrieve the digitized samples. As I progressed along the data path, I found out new and interesting “objects” or pieces of hardware along the way. I encountered a lot of FIFOs in the project (Another technical side note, FIFO or First In First Out is an actual piece of hardware! Go check it out.) and I asked my mentor in frustration “What’s the use of so many FIFOs?” and he said, “You will understand their use in time”. And surely I did, as FIFOs are inevitable in digital system design, they are used for clock domain crossing, scaling data input width, or just as buffers.

The second phase of my internship and the one which I believe is a major chunk of my contribution to the project was to come up with an architecture of generic de-mapper for the JESD204C interface (Yet another side note, JESD204C interface is a serial interface so it packs the sample data at TX end, it is our responsibility to unpack the data at RX end hence the de-mapper). This is where my mentor allowed me to be creative. I bombarded him with ideas I thought were practical. And after multiple rounds of discussions with the team and going back and forth on key parameters of the architecture we finally froze on one.

In the third phase of my internship, I had to implement the architecture on the FPGA board and test the same for functionality. This required me to build upon a project I had already traced out. I worked on Xilinx Vivado software, built all modules, and ran simulation checks for testing functionality. Throughout the process, my mentor helped me with all issues I ran into (Trust me there were a lot of them ). I remember how another member of the lab Mr. Aniket spent the weekend helping me with the timing issues I was facing.

Apart from the regular workflow, I got the opportunity to attend roundabouts with leaders at TI such as Mr Santhosh Kumar MD at TI India and learnt a lot from the experience they shared. I also miss the fun-filled activities I had on weekends with other interns. The HR team at TI really ensured that we got our share of fun and made connections something which all of us were longing for.

In all, the entire internship process was challenging and educative. I am thankful to my mentor for the smooth introduction to the entire project and regular feedback. He empowered me to bring the best ideas I could and contribute to the project. The discussions I had with other members of the lab enlightened me about key areas that I’ve never thought of, helping us create a more efficient design. The best part was being allowed creative freedom to solve a problem that I personally miss in the regular curriculum at college. I hope to visit the facility in real life and relive the dream someday.

PS: Just some suggestions to the readers who love working on hardware. Try to do as many DIY projects in your spare time as possible, be it for some event or just something you find cool and replicable. Try and hope that it does not work the first time. That’s because in the process of making it work you get a chance to improve your debugging skills, plus you will feel happier when it finally does. Another bonus is that you will have a lot of projects to choose from for your resume. In addition to that, here’s a suggestion that’s especially for first and second-year students. Try to put effort into lab and course projects. If you find a possibility of expanding it to hardware, communicate with the faculty and make it happen. I know it sounds intimidating to add another task to the unending list, but trust me you have a lot more time to do things than what you will have in your third and fourth year. Finally, here are some of the places where you might find inspiration for your next endeavor: Instructables, All About Circuits, Great Scott, ElectroBoom, Nandland, Brian Douglas, Lucid Science, The Thought Emporium, Applied Science.