having just graduated a little over a month ago now, i realized how i sometimes get asked by incoming freshmen or current freshmen/sophomores how i got certain internships. usually in the form of requesting advice or a quick resume review so that they can hopefully also follow in my footsteps. so maybe the best way to answer this is by walking through my entire time at texas a&m and laying out everything i did in hopes that other people can learn from it and do even better. i will note that this is going to sound more of a straightforward explanation of my time at tamu, but it definitely shaped how i think and approached things.
coming into tamu (fall 2022), i didn't know what i was going to do with cs. i was okay at coding towards the end of high school, but that was it. i didn't have any projects, any experience, nothing. i had only ever coded in java before and that was all for class assignments. the intro class at tamu was python (engr 102) which syntactically is a bit simpler than java so i was able to pick it up fairly quickly. i also didn't have to take the required chemistry class and was instead able to take a required cs class which helped me take other classes (that the current cs class was a pre-req for) early on. as time passed, i knew that i had to do some projects to build my resume and then use that to apply to internships. the thing is, i just couldn't wrap my head around doing these random "projects" for the sake of adding them to my resume. so i never ended up getting around to doing projects until the first hackathon i attended which was howdyhack. this is probably around the time when building a project piqued my interest a little bit. my friends (adi, akshay, monte) and i were building with gpt-3 before the chatgpt moment happened in november. the idea was a simple email browser extension which summarized emails along with the ability to extract important info. the product was a website that was created with a react framework but ended up in html/css with the simplest api call that took the entire night to figure out (it was a complete disaster). following the hackathon, the same friends and i reached out to a cs organization that wasn't having any meetings but had a pretty large reputation/name in the past. we were able to convince the president to let us become officers and revive the org (texas a&m computing society - tacs). this was pure luck but also really helpful in terms of gaining more experience. as the semester came to an end, by some even crazier luck, my professor for the python class had asked me to be a teaching assistant for the class the following semester. i got really lucky with both the org and the ta role which i think had some of the most impact on me for "experience".
to emphasize, i still hadn't done any real project on my own or anything yet and didn't really have any plans to either. around late november, i started working on this nfl data bowl project with my friends jeff and rahif since we all enjoyed football and knew we needed to add some sort of experience on our resume. we didn't fully understand what we were doing, but continued working on this even throughout winter break. this was just one of those projects to help boost our resume value to help us get internships. in the middle of winter break, i still remember vividly that akshay had reached out to me to start working on a project with him. i think at first i was really hesitant, but because the barrier to entry was low (chatgpt/llm coding/assistance), it made building anything 100x more exciting. this project was a text to 3d printable file generator where you could say for example "car" and an already existing model would output a point cloud of the car which would then be turned into a file that could be printed in 3d. this might've been the dumbest thing we ever built, but in the moment it was so much fun. this is definitely what sparked a light bulb for me to start building "projects". the reason i put it in quotes is because i still didn't believe in building many things for the purpose of my resume, moreso focusing on a single idea that it would help other people or produce a cool result. at this point akshay had another idea which we saved for an upcoming hackathon (in spring 2023), tamuhack. the vision was deep into this in the education technology space where we wanted to help people create study material on the fly. ai coding was still pretty horrible and so our software wasn't the best, but we were learning a lot. around this time i had also gotten accepted into this hackathon called treehacks (insanely lucky) which was a pretty eye opening experience. i got even more involved in the startup like culture since it was hosted at stanford and had gotten a contact at ycombinator. eventually akshay and i reached out to this contact about what we were working on (it was called peer/peeredu) and got told it was a tarpit idea, which essentially means that it sounds good on paper, but actually isn't. this discouraged us quite a bit and made us pivot.
although we pivoted, reflecting back on it, with a bit better execution, it was still a project to continue working on based on the success of similar startups i've seen come and go. from this point we thought about genuine struggles people faced (still within the edtech space) and stumbled upon improving in-classroom experiences. we got accepted into this nsf i-corps program which lets you do customer discovery, build and iterate, and gives a grant at the end. long story short, the pivot to the in classroom experience was focused on helping students learn python in the intro python class. i noticed as a ta that a lot of people struggled because it was a required class for etam, but a lot of people didn't need to ever use it. so they didn't pay attention in class or if they did it would go in one ear and out the other which presents a genuine problem to how coding is taught in general. around this time, i had been given an internship position for "quality engineering" at lockheed martin which i accepted as it was the only internship offer i received. i was also in the interview process with a startup in sf which i got to the final stage of and didn't do so well on and got rejected. everything would probably be a lot different had i received an offer from them but who knows. after receiving the grant we continued to build this throughout all of summer. i would work 4 days a week at my internship, come home the same day and then work on the project friday - sunday in a coworking space. it didn't really feel like i was "working" on anything (more like i was solving something), it was all purely because i believed the problem existed at my core and i wanted to solve it for others.
i was able to convince the professor i was a ta for to let us test our tool the following semester (fall 2023) to see how much it would help people. working with a&m was very difficult here. because we were just students there wasn't a lot of trust in the security of our software (in hindsight, rightfully so). we were pretty annoyed since some of the people we were working with didn't take us seriously and with the process of getting approved taking close to a year or even longer. we also ran into a number of bugs every class period we went to try and use it which caused students in the class to not use it. it wasn't the best outcome, but i know i also gained a sense of complacency after getting an internship offer for microsoft (for the following summer). i quickly realized that all that hard work i put in paid off with what i guess i was ultimately chasing (assuming that the project i was working on wouldn't work out). there was obviously a sense of disappointment that the project didn't work out, but because i learned a lot from it and i knew how difficult the university would be to work with, it wasn't the worst result. to be completely transparent, the internship interview opportunity and offer was purely from luck. i hadn't gotten any referrals or emailed anyone, i just got lucky my resume handed infront of the right manager. not to say that cold emails don't work (b/c i've seen them work), but my experience here was by pure chance. switching gears, this was around the same time i knew i finally had the chance to be consistent in the gym. after grinding academically for pretty much a year straight, i knew it was time to take care of myself. i started working out every day and have stayed consistent since nov. 2023 until now. during that same fall semester i actually got pretty lucky with interviews and got chances to interview with roblox, meta, and some other companies but since i already had an offer locked in, it wasn't worth preparing/trying for them so they didn't end up working out. but i'm adding that here just to give context that getting interviews at these companies isn't impossible just because you go to tamu. not working on the project kind of worked out in my favor since tacs started picking up and the following semester (spring 2024) would prove to be really difficult academically. i took csce 313 honors (computer systems), csce 331 (intro to software eng.), and csce 430 (competitive programming) that semester. now these courses aren't particularly difficult, it was more so the circumstances within the classes. i thought taking the honors version of computer systems would pay off and don't get me wrong, it's valuable but it just wasn't the most intuitive or intriguing and so it proved to be a challenge. there's an interesting story from that class which definitely helped me in the years following, but that's a story for another time. the intro to software eng. class was just outright useless. everything i did in that class, i had already learned and done before through peer/peeredu. the worst part was that i was being forced to do what i enjoyed most with a terrible professor (nicest way i can put it). just a quick tip, don't take this professor (even if it's the only option available). it'll save you unimaginable amounts of time and energy. competitive programming was pretty helpful in preparing for technical interviews, but with the amount of work i had in the other classes piling up from the middle until the end of the semester, there was so much going on i barely learned anything.
following that disaster of a semester, it was finally time for the internship that i wanted all this time. i joined the intune grouping & targeting team. my manager at first was a bit intimidating, my mentor was really kind, but throughout the summer i realized how amazing of a manager i actually had and how i had by far the best mentor ever. from the initial conversations with my manager, i think a big thing that helped me get the internship was that my gpa was a 4.0 at this point. there were a lot of things that i had no idea what i was doing during the internship, but because my mentor was so knowledgeable, i was able to successfully finish my project and a little more all with his help. however, i did slowly realize that although a lot of people idealize big tech, it might not be for some people. i myself found at times that things can and do move slowly relative to working on something of your own. this changed a lot when i returned the following summer but i'll explain later in the timeline. at some point in the middle of the summer i realized that i just still had that urge to build something and so did akshay. we were virtually attempting to build the same thing as cluely before cluely existed where you could have ai running in the background and understand the screen and the audio i/o from you and your interviewer to get past interviews. now obviously building something to cheat with didn't really make a lot of sense if we wanted it to be a product that we could put ourselves behind so there was a bit of a pivot. the reason it worked for cluely is because the founders didn't really care too much and aren't very risk averse. the pivot was similar to a desktop version of claude or gpt before they actually released their desktop versions. we spent that semester (fall 2024) working on this but it didn't really go anywhere. at the end of that semester, i randomly got reached out to interview with nvidia. i think one thing that increased my chance here was by attending some nvidia related virtual meeting in the middle of the semester. it was a simple presentation of a recruiter walking through different roles and opportunities at nvidia as far as i remember, but i think that must've played some role. but again, what i did is something that anyone could do without any sort of referral or cold email. to add to my luck, my team interviewers were amazing and i didn't get asked any leetcode questions; just some simple python knowledge which converted into an offer for fall 2025. also towards the end of that semester akshay and i were dabbling in this idea of lassiprotein and creating a mango lassi protein powder. we actually spent pretty much all of that next semester (spring 2025) working on this protein powder company. the big mistake we made was not putting up an investment ourselves and trying to get capital from others. this is obviously in hindsight as we've seen the interest for protein powder in general grow and more specifically people actually enjoyed a product we made for once. we started posting social media content, tabling at different places, etc all to get a high number of pre-orders.
this even bled into the summer where i returned to intern at microsoft where my experience was a lot different. with the rise in ai agents, i got to work on something entirely separate from the normal processes and build a system to support on call engineers on my team. now i'll be honest i could've done a lot better of a job on what i delivered, but because i controlled most of the flow and changes it was a lot more interesting of an experience than the first time. i was also working with/on the newer technologies so it was more enjoyable. after this internship, i also had my internship/co-op at nvidia over fall 2025. i worked under autonomous vehicles (av) metrics & eval there and it was a very different culture than microsoft. to be completely honest, it did feel like i was shipping code a bit faster which i didn't fully expect coming into it. but for some context, the av org at nvidia is pretty overworked and has to move fast since it's more of a project area whereas the org at microsoft falls under security so reasonably moving slower to ensure security makes sense. although at times it was tough, it was definitely a unique experience and i enjoyed it a lot. towards the very end of the internship, we were finally able to send out orders for lassi protein and people loved it given all the 5 star reviews we had gotten.
after that i had my last semester of college coming up. i wasn't sure where i wanted to work full time or what i wanted to do. the two things on my mind were lassiprotein and exploring the startup scene. i had gotten in touch with a handful of startups and gotten offers from a few of them to join as a new grad. i just wasn't sure if i was making the right decision. around this time, i had also started working on ironbox and so there was a lot going on. i still don't know if i made the right decision and it's something i'll constantly think about because i'm pretty indecisive. only time will tell.
i think that pretty much "summarizes" my time at tamu and the things i did that got me to where i am now. by no means am i at where i want to be, but i think there's so much that can be learned from this that i can now share quite easily. i think being ahead in the cs program, reviving a dead org, becoming a teaching assistant by luck, and then the timing of llms coming out were some of the luckiest things that could've happened to me. along with this, the friends that i had by my side all four years were amazing. i'm sure there's still so much i left out but i think some things can be inferred.
one thing i've noticed is that there's a trend with other people that i've interned with about being teaching assistants and/or ahead in their degree. so if there's one thing you take away from this, it's that you should try to become a ta asap and also be ahead in your degree. the other big takeaway is definitely that there is always going to be luck involved, but you have to put yourself in situations that increase your chances of getting lucky. and nothing i did ever felt forced or boring, every single thing i voluntarily did was through pure enjoyment.
some other practical tips i have are to use jake's resume template to write your resume. have a personal website that you personalize to what you want to show other people so that your personality can shine. it's okay if you suck at leetcode questions because i do too, but have gotten lucky with a lot of technical interviews focusing on my coding knowledge rather than leetcode questions. this is because a lot of questions are focused on hashmaps/dictionaries. on your resume make sure you talk more about what you actually built technically rather than the idea that you built. oh and the single most important thing you can do during an interview is connecting with the interviewer on a really personal level. by doing this, you basically become their friend and a lot of times they likely forget about how you perform on the technical questions.
but just remember to always have fun, know that it's going to take some luck, and only do something if you truly enjoy it.
- aayush