Podcast
From Analytics to Production ML: Team Building, Experiments, MLOps & Fraud Detection
Open original DataTalks.Club episode
From Analytics to Production ML: Team Building, Experiments, MLOps & Fraud Detection
Original Episode
Use these links for the canonical episode and media sources.
- Open the original DataTalks.Club podcast page
- Watch on YouTube
- Listen on Spotify
- Listen on Apple Podcasts
Episode Overview
How do teams move beyond dashboards to reliable production ML—while organizing people, running experiments, and tackling use cases like fraud detection? In this episode Rishabh Bhargava (7+ years in analytics and ML, former Sales Engineering lead at Datacoral—acquired by Cloudera—and early Primer.ai engineer; MS CS Stanford) walks through the practical bridge from analytics to ML in production.
People
Use these links to connect the episode to guest notes.
Chapter Summary
Use these checkpoints to decide whether to open the source transcript.
- 0:00 - Episode Introduction & Guest Overview
- 2:08 - Career Path: Data Infrastructure and Stanford ML Background
- 3:55 - Sales Engineering: Demos, POCs and Data Integration
- 5:35 - Early Machine Learning Work: NLP, Summarization and Entity Extraction
- 6:46 - Prescriptive vs Predictive Analytics: Definitions and Business Use Cases
- 9:32 - Terminology Problems: The Ambiguity of “Data Science”
- 10:48 - ML vs Analytics: Different Goals, Shared Data Infrastructure
- 13:48 - Machine Learning Day-to-Day: Models, APIs, Predictions and SLAs
- 17:38 - Fraud Detection: From Rule-Based Systems to Machine Learning
- 18:39 - Analyst Responsibilities: Dashboards, Reports and Ad-hoc Queries
- 24:23 - Domain Expertise: Analysts’’ Tribal Knowledge and SQL Proficiency
- 26:33 - Documentation Limitations and Attempts to Improve Knowledge Sharing
- 28:42 - Experimental Workflows: Model Experiments, A/B Testing and Shadow Mode
- 31:19 - Experiment Analysis: Segmentation, Uplift and Root Cause Investigation
- 33:30 - Overlaps and Differences: Data Quality, Timescales and Outputs
- 39:04 - Bridging Roles: Notebooks, SQL+Python Workflows and Analytics Engineering
- 41:13 - Investment Trends: ML Hype, Analytics Underspend and Data Infrastructure
- 43:02 - Hiring Imbalance: Prioritizing Data Scientists vs Data Analysts
- 49:01 - Team Organization: Embedded Data Roles Versus Centralized Structures
- 55:41 - Building a Data Team: Hire Data Engineers, Then Analysts, Then DS
- 58:19 - MLOpsRoundup Newsletter: ML Production, MLOps Insights and Resources