Sudarshan Seshadri on High-Stakes AI Systems and the Cost of Getting It Wrong
.svg)
Alto Pharmacy
Sudarshan Seshadri is the Vice President of AI, Data Science, and Foundations Engineering at Alto Pharmacy. Since joining Alto in 2020, he has led efforts to apply machine learning and large language models to high-stakes healthcare workflows, focusing on safety, trust, and operational impact.
Prior to Alto, Sudarshan spent several years at Groupon, where he held senior leadership roles in data science and supply chain optimization. His earlier career includes roles at Mu Sigma and Oracle, and he has consistently focused on using analytics to improve business performance and customer outcomes. He holds a Bachelor of Technology in Mechanical Engineering from the National Institute of Technology Karnataka, and a Master’s degree in Operations Research and Industrial Engineering from Texas A&M University.
Delphina
Hugo Bowne-Anderson is an independent data and AI consultant with extensive experience in the tech industry. He is the host of the industry podcast Vanishing Gradients, a podcast exploring developments in data science and AI. Previously, Hugo served as Head of Developer Relations at Outerbounds and held roles at Coiled and DataCamp, where his work in data science education reached over 3 million learners. He has taught at Yale University, Cold Spring Harbor Laboratory, and conferences like SciPy and PyCon, and is a passionate advocate for democratizing data skills and open-source tools.
Key Quotes
Key Takeaways
Data leadership is decision leadership.
Suddu argues that the core job of a data leader isn’t managing models or dashboards—it’s shaping how critical decisions get made, especially when stakes are high and judgment matters.
AI in healthcare isn’t about speed—it’s about trust.
At Alto, automation doesn’t replace humans; it supports pharmacists with safe, explainable, and regulation-aware systems. Trust is earned through consistency, not cleverness.
LLMs are tools for structure, not generation.
Rather than generating content, Alto uses LLMs to extract, classify, and interpret clinical data—feeding structured signals into downstream decision systems built for precision.
The metrics you track shape the outcomes you get.
Suddu shows how shared, causal, and actionable metrics—like Alto’s “perfect prescription score”—can bridge teams and move the needle on both patient experience and operational performance.
Full-stack practitioners thrive in strong systems.
While the team includes specialists, Alto’s strength comes from people who can carry problems from concept to resolution—and from a culture that supports collective growth.
Judgment scales through structure, not speed.
With thousands of contextual decisions happening daily, Alto invests in infrastructure to scale pharmacist judgment—not just throughput. Probabilistic reasoning and human-in-the-loop systems are essential.
Irreversible decisions demand better tooling.
The systems Suddu builds don’t just support workflows—they influence decisions with real consequences. That’s why rigor, feedback loops, and explainability are baked in from the start.
The next leap for data leaders is executive.
Looking ahead, Suddu sees modern data leaders stepping into broader roles—defined by storytelling, strategic clarity, and long-term decision accuracy, not just technical expertise.
You can read the full transcript here.
00:00 Introduction to Decision Making in Data Leadership
01:08 Challenges in Data Leadership
01:22 Interview with Sudu: Journey and Insights
01:36 Building Trust in AI Systems
01:46 From Bottlenecks to Backbones
02:08 Call to Action
02:20 Introduction to Delphia and High Signal
02:56 Frameworks for Decision Muscle Memory
03:29 Sudu's Career Journey
05:33 Joining Alto Pharmacy
08:03 Building a Data-Driven Culture at Alto
12:49 Evolution of Data Teams: 2012 vs. 2025
16:19 Creating Shared Definitions and Metrics
21:35 Retaining Talent and Building Strong Teams
25:46 Role Design: Full Stack Practitioners vs. Specialists
28:17 The Role of Full Stack Practitioners in Data Teams
28:56 AI and LLMs in Pharmacy Decision Support
29:19 Challenges in Pharmacy Workflow
31:48 Machine Learning Applications at Alto
34:43 AI Pharmacist Assistant: Reducing Burnout
39:41 Using LLMs for Contextual Learning in Healthcare
44:19 Balancing Automation with Safety and Compliance
48:11 Metrics for Measuring AI Systems in Healthcare
51:52 Future of Data Leaders in Executive Roles
55:15 Retrospective Advice for Data Leaders
57:58 Conclusion and Final Thoughts
Links From The Show
Transcript
In the spotlight: Our most popular episodes
Eoin O'Mahony on Why Good Metrics Still Lead to Bad Decisions — and How to Fix It
Eoin O'Mahony—data science partner at Lightspeed, former Uber science lead, and one of the early architects of the system that kept NYC’s Citi Bikes available across the city—argues that positive metrics are meaningless if you don’t understand the mechanism behind them. At Uber, he was careful to make sure his launches both looked good on paper and made sense in practice. Now in venture, he’s applying that same rigor to unstructured data—using GenAI to scale a kind of work that’s long resisted systematization.
.png)
Listen up: Our latest discussions
Hear the hottest takes on data science and AI.
%20(1).jpg)
Get the latest episodes in your inbox
Never miss an episode of High Signal by signing up for the Delphina newsletter.