Uploaded August 2026 | Updated November 2026, 2 days ago
This talk was recorded at NDC Toronto in Toronto, Canada. #ndctoronto #ndcconferences #developer #softwaredeveloper
Attend the next NDC conference near you:
ndcconferences.com
ndctoronto.com
Subscribe to our YouTube channel and learn every day:
/ @NDC
Follow our Social Media!
facebook.com/ndcconferences
twitter.com/NDC_Conferences
instagram.com/ndc_conferences
#machinelearning #bigdata
Building an accurate machine learning model is a challenge - but building a model that succeeds in the real world is even harder. After almost a decade of designing and deploying machine learning solutions across industries one lesson is clear - it takes more than an accurate model to be successful.
This talk unpacks reasons why even well-designed, statistically sound models can struggle. It highlights real-world examples where promising models failed to have the impact expected based on performance in the development phase - covering topics such as organizational pitfalls, poor monitoring and more.
Attendees will leave with practical strategies for scoping ML solutions that are resilient, maintainable, and capable of delivering consistent value in production.
This talk was recorded at NDC Toronto in Toronto, Canada. #ndctoronto #ndcconferences #developer #softwaredeveloper
Attend the next NDC conference near you:
ndcconferences.com
ndctoronto.com
Subscribe to our YouTube channel and learn every day:
/ @NDC
Follow our Social Media!
facebook.com/ndcconferences
twitter.com/NDC_Conferences
instagram.com/ndc_conferences
#machinelearning #bigdata
Building an accurate machine learning model is a challenge - but building a model that succeeds in the real world is even harder. After almost a decade of designing and deploying machine learning solutions across industries one lesson is clear - it takes more than an accurate model to be successful.
This talk unpacks reasons why even well-designed, statistically sound models can struggle. It highlights real-world examples where promising models failed to have the impact expected based on performance in the development phase - covering topics such as organizational pitfalls, poor monitoring and more.
Attendees will leave with practical strategies for scoping ML solutions that are resilient, maintainable, and capable of delivering consistent value in production.










