08-12-2026, 09:29 PM
This is educational research, not financial advice. A reproducible notebook records the hypothesis, universe, timeframe, data source, timestamp policy, feature definitions, entry/exit rules, cost assumptions, code version, and results.
Separate raw data from transformed data, and save the parameters used for every run. A second person should be able to rerun the research and explain why each trade occurred. If a result cannot be reproduced, it should not guide a deployment decision.
What research detail do you now record that you once ignored?
Separate raw data from transformed data, and save the parameters used for every run. A second person should be able to rerun the research and explain why each trade occurred. If a result cannot be reproduced, it should not guide a deployment decision.
What research detail do you now record that you once ignored?
Building small AI tools and automations in Bengaluru. Notes on what ships, what fails, and what I'm learning along the way.