- Published on
Why session-by-session chat is a trap for data science work, and how to layer repo rules, user memory, and a knowledge vault so AI workflows compound instead of resetting every tab.
Data-science
All Posts
- machine-learning (14)
- programming (8)
- python (5)
- architecture (4)
- ai (3)
data-science (3)
- software-engineering (3)
- interpretability (3)
- fundamentals (3)
- dependency (2)
- infrastructure (2)
- pandas (2)
- mathematics (2)
- nlp (2)
- devops (2)
- statistics (2)
- packagemanager (1)
- pip (1)
- anaconda (1)
- workflow (1)
- engineering (1)
- cloud-computing (1)
- startup (1)
- dbt (1)
- data-cleaning (1)
- package-manager (1)
- container (1)
- docker (1)
- probabilistic-programming (1)
- ops (1)
- tensorflow (1)
- huggingface (1)
- ensemble (1)
- Published on
dbt is more than a tool for engineers. It’s a framework for building reliable, production-grade data pipelines—a critical skill for the modern data scientist.- Published on
This blog post provides practical tips and tricks for data cleaning, covering methods to handle missing values, standardize formats, remove duplicates, deal with outliers, clean text, and automate the cleaning process. It emphasizes the importance of proper data cleaning as a foundation for meaningful data analysis.