My goal at Zootera was to build a prototype for a business intelligence sustainability platform. This platform will help businesses find and prioritize nature conservation opportunities to drive more funding towards protecting the environment. My roles and responsibilities included the following:
- Crawl online sources to capture and track project and organizational data
- Clean data model for data from multiple sources
- Build, monitor, and maintain data extraction, transformation, and loading (ETL) process
- Establish, maintain, and document high data quality standards in our databases
- Investigate and address data anomalies, process gaps, and implement solutions
- Consider business problems and needs to provide data that makes an impact

I worked directly with the CEO, and my work was critical in developing and refining the platform, which will allow Zooterra to tap into a new customer target. More importantly, the work also helped create a more encompassing database of projects that provides more visibility into initiatives that require funding and start bridging the $100 billion gap in conservation funding.On a personal level, this role was an amalgamation of learning as well as applying new skills and techniques in the domain of big data and analytics. I was able to use the skills learned at Columbia in a professional setting. I learned how to communicate with supervisors from non-technical backgrounds and how to manage expectations.

The two months were a steep learning curve where I used the wealth of knowledge gained at Columbia, particularly through courses like data analytics, business analytics, and tools for analytics. I used the programming skills acquired from these courses to scrape public data and build pipelines for the extraction of tables and relevant information. I also learned at Columbia (and applied extensively throughout the internship) how to write clean documentation of technical work for a non-technical audience. I applied my learnings from analytics on the cloud to scale up processes and automate different components. I honed my soft skills tremendously by interacting with people from diverse backgrounds. The wide variety of courses and on-campus jobs at Columbia taught me how to compartmentalize, multitask, and prioritize my work at Zooterra.

The challenges I faced were mostly technical — around automating, scaling and speeding up processes. One of the major challenges was to extract paragraphs in order from sustainability PDF reports. The team spent long hours trying to figure out a logic that would be universal for the reports, but without any success. As the deadline was approaching, I suggested that we use the Adobe software to first convert the reports into a generic excel file and then garner the paragraphs from this excel file. This method proved useful as the third party software (Adobe) was able to keep the order of paragraphs intact across reports, and this was a major roadblock while extracting text directly from the PDFs. Another challenge that the team faced was speeding up the process of downloading reports from the web. It took around two minutes to download a single report and we wanted to download reports for 500 companies. To speed up the process, we used the concept of multi-processing, which uses all the cores of the system. We set up the pipeline for multi-processing on Google Cloud Platform and were able to harness the power of 16 cores in parallel.

Key takeaways from this internship are the art of managing expectations and setting deadlines. I also learned how to negotiate, build end-to-end products, and deliver a minimum viable product (MVP) in a fixed timeline. I learned how to collaborate in a cross functional team and work in parallel to maximize efficiency. This experience would make me a good fit for a product manager role. I also learned about the sustainability space and the efforts firms take to reduce their carbon footprint. Another takeaway would be to think about a product from different perspectives and designing products that are user-friendly.