Screening — Priyanka Rao
Machine Learning Engineer · stage screening · fit 82/100
Screening processⓘ
clear✍️ Written questionnaireAsynchronous✨ AI-assessed⚖️ Recruiter-decidedAnalysed
- ✓InvitedJun 26, 2026
- ✓Responses3 / 3
- ✓AI assessment41/100
- ⚖️Recruiter decisionawaiting
💬 Candidate responses
3 of 3 answeredQ1. Describe your experience with machine learning.
52
I built recommendation models at scale using Python and TensorFlow, owning the full lifecycle from feature engineering to serving, improving click-through by 18 percent.
✨ AIMatched skills: Python. 1 quantified result(s); 24 words.
Q2. How do you evaluate a model in production?
35
I monitor production models with offline and online metrics, track drift on input distributions, run shadow deployments and A/B tests before full rollout.
✨ AIMatched skills: none. 1 technical specifics; 23 words.
Q3. Tell us about a hard ML problem you solved.
36
We had a model degrading silently due to a feature pipeline bug; I added data validation, alerting on feature distributions, and a rollback path.
✨ AIMatched skills: none. 1 technical specifics; 24 words.
✨ AI assessment
recommends — never auto-decides41/100
3 answer(s) reviewed scored by automated assessment. Average 41/100. No integrity concerns.
🛡️ Integrity check clear.
⚖️ Recruiter decision
the human owns the call⏳ Awaiting recruiter decision — the AI has scored this screening. Advance or reject below.
Activity history
advanceinterview → screening· Invited to one-way screening
Jun 5, 2026 · 07:33:43