Women in Data Engineering Fueling Atlas AI: Shruti Jain

Anais Tadlaoui


At Atlas AI, we believe that to fulfill our mission of improving lives by enabling more resource allocation in more vulnerable parts of the planet, we need to put together a truly diverse team. As we grow rapidly across the globe, we are committed to do more than put diversity on a mission statement. We’re thrilled to have five women to date on our data engineering team who are driving our innovation and projects forward at Atlas AI. To celebrate them and in hopes of inspiring other women in this field, we are sharing a little bit of their story with you.

Introducing Shruti Jain

Shruti has been working at Atlas AI as a Geospatial Data Scientist for 2 years now. She manages our agricultural pipelines which includes land cover, crop type and crop yield mapping. She also contributes to our service contracts with the World Bank on the 50x2030 initiative and with the Ministry of Agriculture in Kenya. Before joining Atlas AI, Shruti was working at UC Berkeley's Global Policy Lab on mapping the history of economic and natural resources in Africa, Asia, Caribbean and the Pacific using modern satellite imagery and 1.6 million historical aerial photographs. She also has experience working in the development sector in India with Teach For India to teach elementary school students in Delhi, with a startup in the Ed tech space to deliver high quality curriculum to tier 2 towns in India and with J-Pal South Asia to evaluate the impact of rice fortification schemes in rural Tamil Nadu. She went to IIT Roorkee for her Bachelors in Electronics and Communication Engineering and to UC Berkeley for her Masters in Public Policy.

What is your scientific background?

I started in electronics and communication engineering, then went in the development sector as a teacher for 4th and 5th graders at Teach for India in under-resourced schools. I came to the US to do a Masters for Public Policy at UC Berkeley and discovered remote sensing and spatial data with professor Solomon Hsiang. I did research with this professor and stayed at his lab after graduating. 

Which topic are you working on at the moment? Why did you choose this topic and how do you think you’ll make a difference?

I’m working on agriculture, specifically on crop type and crop yield mapping. I’m also working on agriculture service contracts on the 50x2030 initiative at the World Bank and the NSF grant. I believe our work and the initiatives we are contributing to can have a big impact in food security in Sub-Saharan Africa where the data is unavailable. Our datasets can help countries plan ahead to let them know how much they’re producing of a particular crop type. For instance, when COVID-19 first hit, we worked with the Kenya Food Security War Room to help them identify how much maize they had produced in 2019 to plan ahead for the pandemic. Another impactful use case is in the case of interventions. For instance, when seeds are distributed, we can measure if those seeds have a positive or negative impact at a 10m x 10m resolution.

What is a typical day like for you?

I spend about half of my time coding and the rest in meetings or working on presentations for our service contracts. When I’m coding, I’m using Google Earth Engine and Python, large geospatial datasets and cloud computing to process at faster speeds and avoid our computers crashing from processing large amounts of data locally. I also meet with my Applied Data Science team, the founders and in one-on-ones to discuss technical topics, to ask for feedback or advice and lastly in joint client meetings to present findings. 

What are the hardest parts related to this work?

There’s a big learning curve with all of the tools and methodologies but it’s exciting! Another challenge is the uncertainty. When we asses new projects, we must first define its scope and determine what is possible and what isn’t, for instance, to develop a map of intercropping in a given region. Sometimes, that can be difficult to determine as there may be new methodologies we’re incorporating I don’t know yet.

What’s something you recently learned in the field of Data Engineering or Sciences in general you didn’t know before?

I learned how to use CNNs (Convolutional Neural Networks), it’s a machine learning model. Our asset wealth and spending datasets use this method versus a linear regression or Random Forest models.

What’s the most interesting mind blowing thing you’ve learned in Data Engineering that’s really impacted how you perceive the world?

I didn’t think any of the things we are doing were possible, using satellite imagery to assess wealth, the ability to capture patterns and predict spending capacity or using satellite imagery to produce crop yield.

How did you persevere in a men dominated field?

I have never faced explicit bias against me, I have been lucky to be part of respectful teams. However, I find it hard to feel confident and to speak up.

Which changes are needed in Data Engineering to be more inviting to women to JOIN and STAY in this field?

Some things are already changing like consciously thinking about hiring more women in technical roles as well as making a conscious effort to have women in leadership roles to help women engineers feel less like an outlier. Who you see at the top matters.

What is your best piece of advice to women interested in or already in Data Engineering?

Confidence is key. If I could go back in time, I would tell myself to speak up, to apply for that job, to say I’m interested in that new project. My first first reaction was too often that I wasn't competent enough and thinking that someone else could do it better than me. Removing that hesitancy and negative self-talk is important to make sure you're not getting in your own way.

Sign up for our newsletter to get the latest news on investing and growing your business

Subscribe
world map illustration