Published Mar 28, 2023

665: How to be socially impactful and financially successful in your data career — with Josh Wills

Josh Wills delves into the integration of quantitative and qualitative decision-making in organizational settings, drawing from his experiences at Google and Slack. He shares his career journey in data science from management to individual contribution and emphasizes the importance of domain expertise, data quality, and continuous improvement in machine learning engineering.
Episode Highlights
Super Data Science: ML & AI Podcast with Jon Krohn logo

Popular Clips

Episode Highlights

  • Decision-Making

    Balancing quantitative and qualitative approaches in decision-making is crucial for effective data-driven strategies. shares his experience at Slack, where he encountered a culture that prioritized vision and intuition over data, contrasting with his previous work at Google 1. This shift led him to appreciate an integrated perspective, combining metrics with qualitative insights for better decision-making. He cites the Houston Astros as an example of blending data-driven methods with qualitative measures, despite their controversial tactics 1.

    We combine any number of things together in order to come up with the best decision.

    ---

    This approach highlights the importance of using diverse mental models to navigate complex organizational challenges.

       

    Organizational Challenges

    Organizations often face challenges in integrating diverse perspectives, leading to what Josh calls the "infinite loop of sadness." This phenomenon arises when different stakeholders, such as business users, data scientists, and engineers, have conflicting objectives, causing inefficiencies and confusion 2. Josh emphasizes the need for centralized data leadership to mitigate these issues, though he acknowledges the instability of such solutions 2.

    The result of having all these different stakeholders in these different organizations with different objectives, with different goals is kind of this mess.

    ---

    He advocates for rigorous data quality checks throughout the pipeline to ensure reliable outcomes, stressing the importance of immediate detection of quality issues 3.

Related Episodes