Intro
![](images/circle.png)
Sorry for the huge mugshot you just had to see. I wanted to fix its size but got lazy halfway, so I turned it into a feature, not a bug. ;)
TL;DR
I am a data engineer at Cloudflare.
I graduated in 2023 with an MS in Computer Science, in the "Computing Systems" track. After graduation, I spent a year at Regrello, a tech startup simplifying supply chain and manufacturing processes. I shipped many features that our customers needed - I'm especially proud of my work in enabling in-app Search: the backend, the frontend, and the infra.
Previously, I worked at Salesforce for my Summer '22 SWE internship. I was (naturally) super psyched to work out of the tallest building in the west coast.
At Salesforce, I worked on two cool features that users of CRM Analytics presumably really wanted: to pin and to hide their collections. It was cool how my manager Aditya Kochhar and mentor Ashton Vaz let me come up with my own designs, grill review them with the broader team, and implement them fullstack!
Before this, I spent 3 years working at a Silicon Valley company revolutionizing automobile insurtech. This was right after I graduated in 2018 with a BTech in Computer Science from Vellore Institute of Technology, Vellore, India.
Work stuff
Between 2018-2021, I worked at Zendrive- a data-driven insurtech company which crunches insane amounts of driving data to figure out how well you drive your car, using just your mobile phone sensors. The SDK we ship can also figure out if you got into a car crash.
At Zendrive, I got to try on many hats. As an iOS Developer, I built an app "SensorLab" that could record, label, and upload sensor data, while coordinating with other devices in real time. Fun fact: this app was used in car crash tests at BMW facilities, Munich! Not so fun fact: I nearly shipped a wrong build of the app and learnt the wisdom of versioning the hard way.
![](images/sensorlab.png)
As a Data Engineer, I got to design and automate quite a few critical ETL pipelines that fed into the downstream machine learning models. This involved dealing with the complexities of nested data structures, and sheer scale (pre-covid, we were ingesting 5M trips every day). I am proud of my work on scaling up geospatial operations by migrating from PostGIS to GeoSpark. Other than my OKR duties, I trained freshers in working in distributed systems, reviewed their code, and fixed their git issues. We've all been there. For a while, I was also the Airflow guy, helping other teams automate their pipelines as well.
While working with the Backend team, I owned a dashboard that our clients could use to get a bird's eye view on their fleet performance. I started with a design review elaborately covering data flow, costs, optimizations, limitations, and tradeoffs - which my manager Roop loved (his words, not mine). While implementing it, I used many AWS technologies: S3 for storage, Elastic Bean Stalk for orchestration, Route 53 for DNS, and so on. For a denser summary, check out my humble resume.
Non-work stuff
The coolest project I built was a pipeline which can - at the click of a button - download the K most popular posts from Reddit, convert them to images, and post them to Instagram. There were many interesting things under the hood like deduplication (what if the same post comes up the second time you run the pipeline), exploiting GSheets (as a DB, but not recommended), and dynamic text sizing. Sneaked some NLP in there too to infer tags, but the cooler NLP thing I was a part of is this paper that got published in EACL.
Other than all that jazz, I am a trivia buff who made pocket money in his undergrad years by winning trivia quizzes. My YouTube homepage ranges from world history to soccer, k-pop to philosophy. I used to write more earlier though now it is mostly confined to my personal diary (ask me how it dramatically improved my quality of life). I love cats. Of all shapes, sizes, and personalities.
If you've read this far, can I trust you with the most questionable thing I've done?
And this is a blog I write about my journey as an immigrant navigating the waters of USA. It earned me a first date with my now girlfriend!