Why Lane Leveled Up His Data Science Career with Lighthouse Labs

Systems analyst Lane Clark wanted to move up to more technical data roles at BC Public Service, but lacking the right technical skills and experience, he found himself in a holding pattern. After dabbling in self-teaching, Lane took a leave of absence from work to enroll in the remote immersive 12-week Data Science Bootcamp at Lighthouse Labs. Over the course of the bootcamp, Lane gained the knowledge he needed to level up his career (and his salary!), and discovered a whole new passion for machine learning and deep learning. Lane gives us a glimpse into how Lighthouse Labs supports their remote cohorts and teaches the technical and soft skills expected in today’s data science industry.

What inspired you to enroll in a data science bootcamp?

I have a degree in sociology from the University of Guelph, and I have always been involved with computers. After graduating from University, I worked as a technical support specialist for a high school in Montreal, and I took some part-time courses in computer science — that's where I first used SQL. When I moved to British Columbia, I worked as a Service Desk Analyst for the Province of British Columbia and that’s where I got my feet wet with more interesting data-related subjects. I liked the data aspect so much, I decided to take three courses at the University of Victoria in Business Intelligence and Data Analytics to learn more. That’s when I decided I wanted a career in data science. 

At work, I was fortunate enough to get a job-share for a couple of hours every day with the analytics team within my division. I did okay in the soft skills department, but I didn't have the programming experience I needed to really level up. I began researching bootcamps in Canada like BrainStation and Lighthouse Labs, and I originally thought I would enroll in the Web Development at Lighthouse Labs because it was offered remotely. Because of the COVID-19 pandemic, the Lighthouse Labs Data Science Bootcamp became available remotely as well. I immediately pivoted to enroll in that program.

Why did you choose to enroll in a data science bootcamp instead of self-teaching?

I have previously self-taught with online, free Python courses. But with no immediate practical use for this skill, whenever I got 3-5 weeks into the course, I would lose interest and fall off track. I needed an immersive environment. After that experience, I knew I would learn better in an immersive program. 

What was the admissions process like for the data science bootcamp? Did you have to complete a technical challenge? 

Lighthouse Labs gives you a 14-hour Intro to Programming module on their online learning platform called Compass. If you have a background in programming, this module will not take you that long to complete. Prospective students like me use it to prepare for the 15-question technical test, which covers logic questions, programming concepts, and other fundamentals. There was only a short time to answer as many questions as possible in the technical test. Most people won’t answer all the questions, but that was okay.

Were you able to work full-time while completing the bootcamp?

When I started the bootcamp, I was able to take a 3-month leave of absence from my work to fully immerse myself in the bootcamp. A week after I finished the bootcamp, I was in a different scenario than my cohort mates because I was able to go right back to work as a Systems Analyst, and I'm thankful for that. 

I wasn’t able to qualify for a scholarship since I was technically still employed, but some of my cohort members received the COVID-19 Relief Scholarship.

What was a typical day like in the remote data science bootcamp? 

My typical day started at 8am, when I did my prep work for the lecture or caught up on my assignments. From 10am to 12pm, we had lectures. Lighthouse Labs gives estimates for how long things should take us to complete, and we had 6 to 10 hours of work to do after every lecture. How long it takes to complete that work varies on the student. Someone with more understanding of the subject could take half the time to complete the work. As the weeks went on, I found I was staying up later to complete the work and then waking up a little closer to lecture time — everyone seemed to experience that schedule shift at the end! 

Since you did the Lighthouse Labs Data Science bootcamp remotely, how did you connect with your cohort and instructors?

Lectures were held on Zoom, and there was a Discord and Slack to connect outside of the lecture. With the pandemic and everything remote, I was worried that a remote bootcamp would be an isolating experience, and that there would be no community. Plus, bootcamps are intense, and Lighthouse Labs isn't shy about telling you its program can be challenging. That said, Lighthouse Labs nurtures a wonderful community, and I was fortunate enough to become close friends with some of my cohort. People in my cohort were helping each other all the time and sharing their experiences. Lighthouse Labs has built mentoring into Compass, so you can see which mentors are online and click a request-for-help mentor button. It was easy to get a hold of a mentor when you needed help. Most of the mentors would say, "Even if I'm not online, if you have a question, message me on Slack." When they were online, they would message me right back.

What did the Lighthouse Labs curriculum cover?

For the first four weeks, we learned the fundamentals and every Friday we would have a test to test our understanding of the fundamentals they were covering. After the fourth week, we did more hands-on technical work, but on Fridays, we would still circle back on fundamental concepts. 

• Week 1: We used some Bash scripts, but mostly focused on the fundamentals of Python (like functions and data types). 
• Week 2: We touched on data wrangling, data cleaning, exploratory data analysis, data manipulation, SQL, APIs, Json, XML, and Pandas. We learned about consolidating data sources and how to get data. We heavily used Pandas throughout the course. We would implement things first in Python from scratch, and then use Scikit-learn and other similar packages optimized to do the same thing. We spent a lot of time data cleaning and data wrangling — it's satisfying work. 
• Week 3: We covered data visualization. 
• Week 4: We moved on to the fundamental concepts of machine learning, like dimensionality reduction and simple algorithms. We talked about gradient descent and got math-heavy. If you have a background in calculus and you come that far in the bootcamp, week four will be a breeze. I don't have a background in calculus, so it was a little tricky! 
• Week 5: We went more hands-on and learned about machine learning algorithms, like classic machine learning algorithms, logistic regression, random forest, decision trees, support vector machines, and more statistical things. 
• Week 6: We completed a group midterm project. We had a week to wrangle the data of a huge airline flight database. It's a popular dataset of 15 million rows! The goal was to predict arrival delays based on the information within those tables. This was our first time using all of the skills we learned. 
• Week 7: We talked about data pipelines and building our own data APIs. You can automate things in pipelines as opposed to doing it step by step in a Jupyter Notebook.
• Week 8: We got into deep learning, and diving into the differences between deep learning and machine learning. Most importantly, we covered where we might need one over the other. 
• Week 9: We covered time series. We started predicting things with seasonalities. My cohort focused on data for shopping, stock, and weather trends. 
• Week 10: We looked at recommendation engines and reinforcement learning. This unit inspired many of our final projects! 
• Weeks 11 & 12: These were the last weeks, and we concentrated on our capstone project. Lighthouse Labs provided my cohort with guidance if someone wasn’t sure what to do. When we finished our projects, we presented them to employers during the day of the Demo Day event, and family and friends during the evening.  

What did you build for your data capstone project?

A year before I enrolled with Lighthouse Labs, I created a Twitter developer account out of interest. Initially, I was interested in geo-locating tweets and trying to find trends. Then the pandemic happened and my interests changed. Just before the course started, I wrote a Python script to collect tweets, with the hopes that I could use this data during the course. That ended up working out and I titled my capstone project “The Impact of COVID-19 on the Expressed Sentiment of British Columbians on Twitter.” The script I’d written had been scraping all tweets related to the hashtag "#BCPoli". By the end of the bootcamp, I had over 400,000 tweets from August to November 30th! I used that data in my final analysis for the capstone.

There was statistical value in the project, but I used a sentiment analysis called Vader to calculate sentiment polarity and sentiment scores for the tweets. I mapped them over time against major political events, like the provincial election. Along with that data were a bunch of COVID-19 metrics and statistics from the Government of Canada, such as the daily COVID-19 deaths and case counts. As expected, and as I’d defined in my hypothesis, my project could illustrate a moderately strong correlation between the daily COVID case counts and sentiment on Twitter. 

Which data roles did you feel qualified for after graduating from the bootcamp?

Data science is a challenging field to compete in, but there are many roles that need data literacy and the fundamental principles of data science. The government sector is digitizing everything they do, so they’re using a lot more data. Since graduating, I have applied for a business analysis role within my division, which is actually more like a data analyst/engineering role. Anyone on the data science job hunt now should keep in mind that job titles may be outdated, so be sure to read the job description.

What kind of Career Services assistance did you receive from Lighthouse Labs?

The Career Services team was so nice and super helpful during our meetings. Career Services helped me update my resume and maximize the impact of my LinkedIn profile, which was beneficial to me as an upskiller and not something I would have thought to do on my own. Lighthouse Labs prepared my cohort for interviews. We had three or four mock interviews where we practiced entirely technical data science questions, which was helpful. It's something that I keep in mind while moving forward. 

Have you been able to take on more data responsibilities in your current role at BC Public Service?

Prior to starting the bootcamp, I was employed as a Systems Analyst. I'm back in that role, but now I am working two days a week with the Analytics Team. Even though my title has not changed, my role has shifted immensely. I'm doing more technical things and I’m brought in on more technical projects because I have the skill. I'm putting the skills (especially Python) I learned in the course to good use. I haven't used machine learning or deep learning on the job yet, but I have been hands-on with SQL, Python, and other core concepts relating to data. I wasn't able to do all the work I am doing now before the data science bootcamp. 

The first Python ticket I had after graduating Lighthouse Labs was implementing improvements to the logging functionality and alerts of our cronjobs that run overnight. These tasks collect data from an API or process data and send it somewhere else. Previously, whenever a scheduled task completed, everyone on the development team would get a cryptic email alerting us whether it was completed or not. To improve this process, I wrote a function that collects information on the actions occurring in a script, and reports back to us if the job was completed or not. With my new system in place, we get a catered report of what worked and what didn't, instead of getting the info logs. 

I've been refactoring some SQL to require fewer code changes when we onboard a new site. I also updated a Bash script that queries  Amazon Redshift for table statistics and then writes that data into a new table on Redshift that monitors table sizes and row counts so we can see how the tables change over time more easily.

Do you think enrolling in a bootcamp like Lighthouse Labs could help you get promoted to more mid-level or senior-level roles?

Definitely! I would not have screened into the grid level of the jobs that I'm applying for now if not for Lighthouse Labs. I know because I tried before I enrolled at the bootcamp. Before Lighthouse Labs, it felt like a long shot, and now it doesn't feel that way anymore. My work experience has broadened, which is something that I can leverage for internal promotions as well.

This new knowledge has enabled me to contribute in ways that I could not previously. Before Lighthouse Labs, I couldn't compete for the data roles I was interested in, but now I’m competing for a permanent position on this Analytics Team. Lighthouse Labs helped me get there.

Looking back on 2020, was enrolling in Lighthouse Labs’ Data Science Bootcamp worth it for you? 

Yes, it was definitely worth it. Without Lighthouse Labs, there is no chance that I would have been able to dedicate the time to learn all of this on my own. Since the bootcamp is so immersive, sometimes I was putting in 70 plus hours a week into my studies. I would never have been able to do this if I had chosen to self-teach while still working.

Do you have any tips for anyone considering upskilling while still holding on to their job? 

Knowing it was possible helped me the most. I saw that someone else in government had taken a web development course through Lighthouse Labs and then returned to work. I have a good relationship with my bosses and I didn't feel like it was an imposing question. Since I wasn't able to screen into roles I wanted, a bootcamp would give me the opportunity to expand my skill set and achieve those goals. 

If you can't take off as much time as I did, Lighthouse Labs offers part-time, certificate programs that only take 6 weeks to complete. You get so much out of them. If your employer values you, you owe it to yourself to have that discussion. You want to invest in yourself, and frankly, your employer should want to invest in you, too. 

What do you wish you knew before enrolling in the Data Science Bootcamp at Lighthouse Labs?

More Python! Technically, you don't need to know any Python when coming into the program, but the better you are at Python, the easier it will be for you. All the problems you work on will require Python as your primary tool. 

When I came into the bootcamp, machine learning sounded cool, but I didn't know how much of the course would focus on machine learning and deep learning. Machine learning and deep learning was way cooler than I expected. Even the simple logistic regression was so interesting. After week 5, the data science bootcamp took off and wowed my whole cohort. That's when we realized what is actually possible with data science. From that point on, I was hooked!

Find out more and read  Lighthouse Labs reviews on Course Report. This article was produced by the Course Report team in partnership with Lighthouse Labs.