In this Leading with Data session, meet David Zakkam, a leader with 19+ years of experience. David held key roles at Swiggy, Meta, and Uber, currently serving as Uber’s Director of Data Science. He shares insights on data science’s dynamic role in tackling challenges, optimizing customer experiences, and navigating crises like COVID-19. David’s journey, transitioning and solving complex problems, provides valuable perspectives for data enthusiasts and industry pros.
You can listen to this episode of Leading with Data on popular platforms like Spotify, Google Podcasts, and Apple. Pick your favorite to enjoy the insightful content!
Key Insights from our Conversation with David Zakkam
- Transitioning from consulting to product companies offers a more integrated and impactful role in applying data science to business.
- During crises like COVID-19, data science can play a pivotal role in real-time decision-making and recovery.
- Customizing customer experiences through data-driven insights can significantly enhance engagement and growth.
- Integrity work in social media platforms involves complex, adversarial problems that require constant vigilance and rapid response.
- The future of data science in mobility includes improving customer and driver experiences, integrating services, and leveraging AI for creative solutions.
Now, let’s look at David Zakkam’s responses to the questions asked in the Leading with Data.
How did your journey in data science begin, and what were your early days like?
My professional life can be divided into three distinct phases: 5 formative years, a decade of data science consulting, and the last 5 years in tech companies. I started as a biochemical engineering graduate from IIT Delhi, working on computational biology, which you could think of as data science for biology. Post-MBA, I transitioned into tech, and after a stint in sales, I formally moved into the data science profession.
What was the transition like from consulting at Mu Sigma to working at product-focused companies like Swiggy?
The transition was exhilarating. In consulting, you don’t have the same level of company integration to make impactful changes. In a product company, you’re part of the entire journey, working with various teams to ensure data science is effectively applied to business. The end-to-end ownership brings higher responsibility and satisfaction. My broad experience was invaluable, especially when dealing with complex, unsolved problems.
Can you share an interesting problem you tackled at Swiggy during the COVID-19 lockdown?
When the lockdown hit, Swiggy’s business dropped by over 90% overnight. We formed a 24/7 WhatsApp group with top company executives to address the crisis. We tackled a range of issues, from understanding district-level lockdown interpretations to tracking migration patterns of our workforce, which impacted our market share. These efforts helped us return to pre-COVID levels within six months.
How did Swiggy use data science to optimize customer experience and restaurant growth?
We used analytics to customize coupons based on customer behavior, encouraging them to increase their order value or frequency. For restaurants, we built a tool to simulate and optimize their spend on various promotional options, providing them with actionable insights to grow their business.
What were the challenges and exciting aspects of working on content integrity at Meta?
At Meta, we dealt with various forms of inappropriate content and behavior, from fake accounts to harmful interactions. The integrity team, consisting of thousands of engineers and data scientists, used sophisticated measurement and sampling techniques to improve our classifiers. The challenge was the adversarial nature of the problems, where attackers constantly evolved their tactics, requiring us to be agile and responsive.
What kind of data science problems are you currently working on at Uber?
At Uber, I lead teams focused on mobility growth, new verticals like high-capacity vehicles and rentals, driver and courier quality, and merchant growth on the delivery side. We’re working on enhancing customer and driver experiences, improving reliability, and ensuring seamless integration of services like taxis with Uber’s platform.
What does the future hold for your team at Uber, and what are your thoughts on generative AI?
While the current hiring plans are uncertain, the long-term goal is to grow the data science team in India to match the 30% tech presence. As for generative AI, I see its potential in creative use cases where it can generate meaningful content. However, most business problems today are deterministic and require optimization techniques rather than creativity.
Summing Up
David Zakkam’s data science journey, from computational biology to impactful tech roles, tells a compelling story. His experiences highlight data science’s transformative power in critical business decisions, especially during crises. Navigating Swiggy’s challenges in the COVID-19 lockdown, addressing content integrity at Meta, and leading data-driven solutions at Uber, David’s insights reveal diverse data science applications.
For more engaging sessions on AI, data science, and GenAI, stay tuned with us on Leading with Data.
