Hi, I'm Farah.

I started out trying to understand how the brain decides what to look at. Now I build the data systems that help organizations act on what they know. From modeling human attention at Harvard to engineering data pipelines for coffee farmers in Ethiopia — the thread is the same: making sense of complex signals so better decisions can follow.

I'm a data analyst at Enveritas (May 2022 – present), a nonprofit using technology to lift coffee farmers out of poverty. I work across analysis and engineering: I co-built the core scoring engine used to process field survey data across all client reports, migrated the codebase into a typed framework that reduced errors by 98%, and set up Looker data models for internal and external reporting. I also do analytics projects for roaster clients and teams like operations and impact — including predictive models of fertilizer usage and agricultural yield, and the data work behind the Ethiopia Sustainability Dashboard.

Stack: Python · SQL · Snowflake · LookML · Pandas · Streamlit · R

Before moving into tech, I was on the academic track — working toward a professorship in cognitive science. I completed my Ph.D. in Computer Science at the University of Massachusetts Boston under Marc Pomplun, and went on to a postdoctoral fellowship at the Visual Attention Lab at Harvard Medical School, collaborating with Jeremy Wolfe and Gabriel Kreiman. My research focused on visual attention and memory, using CNNs to classify eye-tracking patterns, building computational models of visual search, human memory and working with behavioral, time-series, and image datasets including COCO.

During the pandemic I made a deliberate pivot — trading the path to professorship for a career in industry where I could apply the same analytical and engineering skills to problems with immediate impact. I hold an M.A. in Psychology and a B.A. in Computer Science and Physics, both from Mount Holyoke College.