Interview Prep · ML Systems Lab

PharmEasy Senior Data Scientist / MLE Interview Guide

PharmEasy operates health-tech across e-pharmacy, diagnostics, and teleconsultation. Senior DS / MLE roles cover demand forecasting, recommendation, fraud (prescription validation), inventory, and clinical analytics. The interview tests healthcare-specific judgment alongside ML depth. This guide covers it.

PharmEasy is one of India's largest health-tech companies, operating across pharmacy, diagnostics, and telehealth. Senior DS / MLE roles work on demand forecasting (SKU-level inventory), recommendation, prescription fraud, doctor-patient matching, and clinical analytics. The interview profile differs from generic e-commerce: healthcare regulations and clinical correctness matter.

The loop structure (Senior DS / MLE)

Standard loop: recruiter screen → analytics case + SQL → ML depth → ML system design → behavioural.

Round-by-round breakdown

Round 1 — Recruiter screen. 30 min.

Round 2 — Analytics case + SQL. 75 minutes. SQL on pharmacy schema (orders, SKUs, refills, prescriptions). Case: "drive incremental refill rates for chronic medication."

Round 3 — ML depth. 75 minutes. Standard ML rigour plus healthcare flavour: time-series forecasting for inventory, classification for prescription validation (OCR + NER), recommendation with safety constraints (no contraindications). Healthcare class imbalance: rare diseases, rare adverse events.

Round 4 — ML system design. 60 minutes. "Design demand forecasting for 50,000 SKUs across 100 warehouses" or "design prescription fraud detection pipeline."

Round 5 — Behavioural. 45 minutes. Healthcare ethics, mistake handling, regulatory compliance.

What PharmEasy weights distinctively

1. Healthcare correctness. A wrong recommendation can hurt a patient. Senior MLEs build with safety constraints natively. 2. Inventory forecasting depth. Pharmacy demand is hierarchical (SKU × warehouse × time), often intermittent, with stockout sensitivity. 3. Prescription processing. OCR + NER + drug-drug interaction validation is a unique technical area. 4. Regulation. Drugs and Cosmetics Act, prescription verification rules.

Top 10 questions PharmEasy senior DS / MLE candidates face

1. "Design demand forecasting for 50,000 SKUs × 100 warehouses. Many SKUs have intermittent demand." 2. "Build a prescription validation pipeline: OCR → drug name extraction → dosage parsing → contraindication check." 3. "Recommendation with safety constraints: never recommend drug A to a patient on drug B due to interaction." 4. "Calibrate a fraud detection model for prescription validation. Wrong rejections frustrate patients; wrong acceptances are unsafe." 5. "Diagnose: refill conversion rate dropped 15% in Mumbai. Approach?" 6. "Hierarchical forecasting — explain reconciliation across SKU, warehouse, region levels." 7. "Intermittent demand: how does Croston's method differ from standard forecasting and when do you use it?" 8. "Healthcare class imbalance: 0.01% adverse event rate. How do you train and validate?" 9. "Doctor-patient matching for telehealth: what's the objective, what are the constraints?" 10. "When would you NOT deploy an ML model in healthcare even if it improves a metric?"

The prep path through MSL

  • Tier 5 (Evaluation) — calibration, class imbalance.
  • Tier 7 (Production Engineering) — all 5 posts.
  • Tier 8 (Monitoring) — all 5 posts.
  • Posts 88 (Time Series), 95 (Anomaly), 98 (Fairness), 125 (Hierarchical Forecasting).
  • Practice: FraudDetectionTab, LoanDefaultTab.
  • Common failure modes

  • Generic ML answers without healthcare correctness layer.
  • Underestimating inventory forecasting complexity.
  • Inability to articulate the safety-vs-accuracy trade-off.
  • Compensation

    PharmEasy Senior DS / MLE in 2026 ranges roughly ₹28 lakh – ₹50 lakh for 5-7 YOE. Lead roles ₹65 lakh+. Below pure-tech rates with strong healthcare-ML depth.

    Continue interactively
    Read this post inside ML Systems Lab — with Simplify toggle, interview Q&As, inline glossary, and the MLE Path forward pointer.
    Open in MSL →