New Data Science Team Lead struggling with aggressive PM on timelines and model expectations
\n \t
\n
<\/BR>
\n
\n \t
\n
<\/BR><\/p>\n
I\u2019m a data scientist who was recently promoted to be a data science team lead. Overall I enjoy the role, but I\u2019m running into a recurring challenge with a very aggressive product manager (also a leader) that I\u2019m not sure how to handle well yet.<\/p>\n
There are two main issues:<\/p>\n
1. Project timelines<\/strong><\/p>\n Whenever we plan a project, she strongly questions why the data science timeline is \u201cso long.\u201d 2. Model performance expectations<\/strong><\/p>\n She also frequently questions why the model performance \u201cisn\u2019t better.\u201d I understand that pushing for faster delivery and better results is part of a PM\u2019s job. I\u2019m not against being challenged. But I\u2019m struggling with:<\/p>\n For those of you who\u2019ve been senior ICs, DS managers, or team leads:<\/p>\n Any advice, examples, or even \u201cthis is normal, here\u2019s how to survive it\u201d stories would be greatly appreciated.<\/p>\n<\/p><\/div>\n submitted by \/u\/Rich-Effect2152 <\/a>
From my perspective, the timeline reflects real uncertainties: data quality issues, iteration cycles, experimentation, validation, and sometimes dependency on upstream systems. But in discussions, it often turns into \u201cwhy can\u2019t this be done faster?\u201d rather than a conversation about trade-offs or risk.<\/p>\n
Even when we\u2019ve already applied reasonable feature engineering, tried multiple models, and are close to what I believe is the practical upper bound given the data, the response is often \u201ccan\u2019t we push it further?\u201d without a clear cost-benefit discussion.<\/p>\n\n
\n
[link]<\/a><\/span> [comments]<\/a><\/span>\n<\/div>\n