Onkur Sen

"The way to be is to do." Confucius

Week 1 at Palantir: Drinking from the hose and putting out fires

May 28, 2013

Ah, the San Francisco Bay Area! It’s great to be here for another summer. This time I’ll be in Palo Alto with Palantir, a platform for organizations to integrate various disparate types of data into a central tool for analysis. This description might sound vague, but Palantir is being used for many organizations and companies that are household names. For example:

As you can tell from the examples above, much of what Palantir does focuses on making a difference in both mundane and extraordinary circumstances. One of their taglines is “tackling problems for a better world”, which is the central focus of the philanthropy team (they were responsible for the first project above). For those of you who know me personally, you’ll know that one of my personal visions is to combine the abstract and the practical into something that can touch and impact people. The philanthropy team seemed like a perfect fit!

My indoc(trination) page said that the first week at Palantir would be a blur. I think a splat on a blank canvas (a la Jackson Pollock) is more appropriate. Although the onboarding process at Palantir is very well structured, there is so much information spread throughout the company being thrown at you at one time that it seems like your mind is stretching at the seams. Our first day started off with a general introduction and the usual employment paperwork and a couple of talks about IT, security, and company culture. After breaking for lunch with my awesome mentor Dan Tse, I spent the rest of Tuesday and all of Wednesday setting up my development environment, a painful but necessary task at the beginning of most internships.

By Thursday, I was finishing up configuration and getting my bearings for my first project. One of the philanthropy team’s main high-urgency tasks is a quick-turnaround response to natural disasters. It just so happened that the night before and the day that I flew out to California, a tornado ripped through Oklahoma and laid waste to the small town of Moore. This was incredibly frightening news for me as I’m from Tulsa, which is less than a couple of hours’ drive away from Moore. However, when I reached Palantir, I found out that they were partnering with Team Rubicon (who have previously worked with Palantir during Hurricane Sandy) to provide disaster relief on the ground to the victims. My first task would be to do a data integration on affected sites into the live Palantir instance by Friday.

Let me reframe the context for that last sentence: starting the third day of my internship, I was expected to push code by the end of the next day on something that would be used in a real-world disaster relief effort close to home. This is definitely not going to be an ordinary summer.

I was driven by adrenaline and a sense of purpose. I had to deliver on this to align with the vision I had set out before I arrived. After cobbling together development environments, scripts, server instances, and phone calls, I had something I thought would be ready to deploy. However, we didn’t have enough time to test locally or on a staging environment, so we did the insane: we pushed it live immediately. This was a particularly risky move as disaster response requires a tactful balance between efficiency of execution and perfection. Luckily, everything went smoothly, and the end result was the map below (I couldn’t even view it; I had to get my colleague to relay the details over phone and then send me this image!). The path of the tornado is constrained by lines in black, and the places that were affected are marked by yellow dots. At the time of writing, Team Rubicon was working on the portion just west of I-35.


This is Palantir for me in a nutshell: work extremely hard (12-hour days for me every day this week) on problems that have immediate practical signifance with people who are smart, helpful, and nice. This doesn’t seem to be the norm, though: I know interns who are either stuck in onboarding or (seeming to) spend their time in “fun activities” more than their actual work. I’m glad I am not among them; I’m here to gain as much as I can from this experience and contribute in ways that are meaningful both to others and to me. However, I do plan to enjoy and explore the Bay Area; Memorial Day weekend will be a much-needed and enjoyable break!

Let the games begin.

See my professional work

Get in touch with me

See my personal creations