Common Mistakes AI, ML & Data Science Interns Make (and How to Avoid Them)
By K. Boopathikumar, CEO & Founder · 16 November 2026 · 4 min read
ML intern mistakes often make results look better than they are. Reviewers spot them quickly.
The mistakes, and what to do instead
- 1. Data leakage between train and test. Instead: split first, then fit preprocessing on training data only.
- 2. Reporting accuracy on imbalanced data. Instead: use precision, recall and F1.
- 3. No baseline. Instead: compare with a simple model or rule.
- 4. Tuning on the test set. Instead: use a validation set or cross-validation.
- 5. Messy, unrunnable notebooks. Instead: restart and run all before sharing.
- 6. Hiding limitations. Instead: write what the model cannot do.
The habit behind all of these
Ask early, commit small, test before you say done, and write down what you did each day. Interns who do these four things get better reviews than interns who are simply faster.
Next steps
- Study the free AI, ML & Data Science learning path and take its module quizzes for a free certificate.
- Follow the AI, ML & Data Science skill track from basic to advanced.
- Do real, reviewed work in a AI, ML & Data Science internship and keep every project in your own GitHub.