5 Pros and 5 Cons of Insight Data Science Fellowship

April 01, 2019

If you’re reading this, you’re likely considering enrolling as an Insight fellow. I just finished the data science track in their Silicon Valley location (aka DS SV 19A), and I’ve thought a lot about if it was the “right choice” for me. As many people on the internet suggest, you could alternatively just get a job and probably start making money sooner. Personally, I’m really glad I had the opportunity to be an Insight fellow, and I would absolutely make the same decision again. That said, this might not be the case for everyone, so below I’m sharing my opinions of both the upsides and downsides. I hope this will provide a more balanced and fair perspective of the Insight program, as the only posts I had found previously were very one-sided.

5 Pros

1. Improved awareness of opportunities in data science

Prior to attending Insight, I thought relatively simplistically about where I wanted to work as a data scientist. I knew I didn’t want to do finance, and I thought it would be cool to work at a place whose product I use, so I applied to Facebook, Yelp, CoffeeMeetsBagel, etc.

I didn’t think much about small/medium size companies, and I honestly wasn’t sure how to navigate that space to figure out which companies would be good fits. But at Insight, I went to about 40 1-hour sessions in which representatives from the company (usually data scientists and their managers) would come in and talk about what it would be like to be a data scientist there. Getting to see all of these presentations and Q&A really strengthened my sense of what I look for in a data scientist role.

2. Personal introductions to companies

Compared to submitting a resume online (even with a referral), the introduction that Insight provides fellows to companies substatially helps our chances of being called for an interview, given that we’ve personally met the team once or twice. Insight also does a fair amount of work to introduce us only to companies that are actively hiring data scientists, whereas I may waste time applying to companies that just have stale job postings. The hiring managers from these companies that work with Insight are often enthusiastic to work with us, and so it’s likely that they’ll hire at least 1 fellow.

3. Alumni mentoring and network

My favorite part of Insight was getting to interact with past fellows who are now data scientists. Alumni mentoring sessions ranged from feedback on our work to mock interviews and general advice about transitioning into a data scientist in the tech world. Alumni also came in to do workshops covering things like product analytics, SQL, and data challenges. Over the course of the program, I met dozens of alumni, including at least 10 who I met with 1-on-1 for at least 30min, and a “tech advisor” who I met with almost 10 times. All that is to say: the alumni mentoring and network here is immensely active and helpful.

4. Fellow fellows

Nevermind, my favorite part of Insight was going through the job interview experience along with ~30 other fellows. It made the project work, studying, and interview prep so much more enjoyable by learning from other people instead of just the internet. The community in the Insight office is great, and I enjoyed going in each day because I got to hang out (and work) with the other fellows. It was certainly a bonding experience that gave us both new friendships and a strong network of data scientists we’ll have when we want to look for jobs in the future.

5. Interview preparation

In addition to spending time with companies, program alumni, and fellows, the “education” part of Insight helps us learn both technical and nontechnical things to succeed in job interviews and figure out what job is best for us. There are tons of workshops covering topics like negotiation, venture capital funding, and behavioral interviews. We also get to meet 1-on-1 with the program directors and other staff throughout the process to help think through whatever we are going through at that time. This was especially helpful after week 7 of the program (the “official” end), in the interviewing stage.

5 Cons

1. Limited selection of companies

Insight might not be a great fit for you if you want to work at very specific companies. An important thing to know is that even though a company has partnered with Insight in the past does not mean that they are recruiting data scientists in your session. It’s also important to know that there are more small- and mid-size companies that come visit Insight compared to large companies. This isn’t necessarily a bad thing, but if your heart is set on working at a big company, maybe you should just apply to them directly.

There are 30-40 companies (about 1 per fellow, depending on the size of the program) in the “first wave” of the program. We meet these companies in the first 4 weeks, present our projects to 6-8 of them during the next 3 weeks, and then start (hopefully) interviewing with them. With this number of companies, there’s decent selection to identify many you would like to work for, as long as you’re not too picky. If none of the first wave companies work out, there are still opportunities rolling in after that, but they are more sparse.

Additionally, even if a company is partnering with Insight, companies progress through interviews at very different speeds (larger companies tend to move slower), which can be an issue…

2. Constrained timeline

Because Insight operates at a very structured timeline, this may not work out in order to get an offer from your favorite opportunity before you need to decide on another offer. You can control this to some extent by delaying some interviews and expediting others, but it doesn’t always work out cleanly. Some of the most interesting opportunities can be extremely slow because they are having bureaucratic issues, whereas the smaller companies move quickly to hire someone as soon as possible. To be fair, this could totally happen if you apply on your own too, but you have a bit less control at Insight.

3. Stress

Insight is straight-up that the 7-week program is “intense.” And I personally found the first 3 weeks very stressful because I was not used to doing a project on a rapid timeline, and there was a lot of stuff to do, and it doesn’t always work as planned. Some people found the interview prep stage more stressful, but from grad school, I had become comfortable with having a never-ending list of things to read and study. I studied lots of nights and weekends, and I think everyone did to some extent, depending on how much in a rush they were in scheduling their interviews. I’m sure glad I didn’t have to worry about raising a family or completing my thesis during Insight, but some people do that perfectly well.

4. Draining savings

I was lucky to have saved money during grad school to support myself during this time, but it was a bit discomforting to see how quickly money goes in SF. If you’re one of the lucky ones, you’ll know someone whose place you can crash at during the program, but some people were living on very tight budgets.

Overall, I think this is a reasonable compromise because I think the preparation that Insight gave me will lead me to earn more in the future, or at least get more satisfaction out of my career.

While some fellows have job offers 10 weeks after starting the program, some fellows will have to live off of savings for several more weeks because so many factors in the interview process are out of our control. I am not sure of the actual numbers, but I would guess that about half the fellows have job offers within 13 weeks of starting the program, and there’s a long tail after that, with some fellows still looking for the right fit after the next session begins.

5. (Slight) competition between fellows

As I said above, being with the other fellows is awesome, but since we’re all going for the same jobs, there’s inevitably some competition. Because. They can’t hire all of us? However, this isn’t terrible because often companies are open to hiring multiple data scientists, and the community at Insight is very supportive. Some have suggested that people may progress farther in the interview processes if they applied on our own instead of alongside several other fellows because we are probably being compared to one another. It’s a valid thought, but I’m not too convinced that comes into play too much.

4.5 stars

I like ratings. So overall, I would give Insight 4.5 stars (out of 5). Note this was my experience as a data science fellow in Silicon Valley, and that each program and location is unique and may have distinct pros and cons. If anyone reading this disagrees or thinks I’m missing something important, send me a message, and I’m interested to hear what you think!