Podcast
Practical Data Science & ML: Feature Engineering, Model Monitoring, Data Governance & Storytelling
Open original DataTalks.Club episode
Practical Data Science & ML: Feature Engineering, Model Monitoring, Data Governance & Storytelling
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 you move from models that look good on paper to reliable machine learning in production—while keeping data clean and stakeholders aligned? In this episode Thom Ives, founder of Integrated Machine Learning & AI and a veteran data scientist, walks through practical approaches to feature engineering, model monitoring, data governance, and data storytelling. Thom draws on a career spanning industry roles and mentoring to contrast concept-focused learning versus specialist detail work, and to explain why.
People
Use these links to connect the episode to guest notes.
Chapter Summary
Use these checkpoints to decide whether to open the source transcript.
- 1:15 - Episode Introduction & Guest Thom Ives
- 1:50 - Concept-focused learning vs. detail specialization
- 3:21 - Career journey: naval nuclear program, grad school, early AI
- 5:11 - Industry roles: HP, ON Semiconductor, SaaS AI work
- 8:52 - Mentoring & community building: integrated mentoring origins
- 9:12 - Why business acumen matters for data professionals
- 10:51 - Role clarity: data scientist versus domain expert
- 13:39 - Rapid delivery & customer-centric feedback (MVP / tracer bullet)
- 19:32 - ETL reliability, data collection gaps, and advocating for clean data
- 21:39 - Shared responsibility: data governance and data literacy
- 23:52 - Data-driven vs. data-informed: definitions and practical balance
- 28:09 - Analytical skills & data storytelling before modeling
- 31:21 - Machine learning development pipeline: feature conditioning to modeling
- 34:54 - Feature scaling, selection, and engineered features for business insight
- 40:46 - Addressing collinearity with PCA and pursuing parsimony
- 45:53 - Model selection: accuracy, variance, and generalizability
- 47:30 - Monitoring models in production: data drift, concept drift, and maintenance
- 49:28 - Essential business skills: explainability, persuasion, and influence
- 50:42 - Relationship-building: informal check-ins, lunch & beer networking
- 55:49 - Remote rapport: virtual lunches, video calls, and building camaraderie
- 58:36 - Integrated ML & AI community: structure, ethos, and free resources
- 1:03:36 - Joining the Slack community and accessing resources