Interview Prep · ML Systems Lab

Cleartax Senior MLE / Data Scientist Interview Guide

Cleartax (now ClearOne) operates in tax compliance for individuals and enterprises. Senior MLE / DS roles cover document processing (OCR + NER for invoices and tax documents), entity extraction, anomaly detection in tax filings, and product analytics. The interview tests document AI and tax-domain reasoning. This guide covers it.

Cleartax (rebranded as ClearOne) runs tax compliance and GST infrastructure for individuals and businesses. Senior MLE / DS roles work on document AI (invoice OCR, GST return parsing), entity extraction (vendors, amounts, tax categories), anomaly detection in returns, and product analytics for enterprise customers.

The loop structure (Senior MLE / DS)

Standard loop: recruiter screen → ML + NLP fundamentals → ML system design → behavioural.

Round-by-round breakdown

Round 1 — Recruiter screen. 30 min.

Round 2 — ML + NLP fundamentals. 75 minutes. Standard ML rigour plus NLP depth: document parsing, NER, entity linking. OCR pipeline design.

Round 3 — ML system design. 75 minutes. "Design invoice OCR + GST classification pipeline" or "design anomaly detection in GST returns."

Round 4 — Behavioural. 45 minutes.

What Cleartax weights distinctively

1. Document AI depth. OCR + layout-aware NER. The unique tech of this domain. 2. Tax-domain expertise. GST categories, HSN codes, tax classification logic. 3. Enterprise reliability. Tax compliance can't be wrong; explainability matters. 4. Bulk processing. Millions of invoices per day for enterprise customers.

Top 10 questions Cleartax senior MLE / DS candidates face

1. "Design invoice OCR pipeline. Latency, accuracy, fallback strategy." 2. "NER for invoice data — vendor, amount, HSN, tax — what model and how to evaluate?" 3. "GST classification — predicting HSN codes from item descriptions. Architecture?" 4. "Anomaly detection in GST returns — what patterns matter?" 5. "Document layout understanding — when do you use LayoutLM vs simpler approaches?" 6. "Active learning for invoice labelling — how do you set this up?" 7. "Diagnose: invoice extraction accuracy dropped 8% in a specific customer segment. Approach?" 8. "Reconciliation between GSTR-1, GSTR-2A, GSTR-3B — how do you architect this?" 9. "Class imbalance in tax category prediction (some HSN codes are rare) — handling?" 10. "Walk through your most impactful project."

The prep path through MSL

  • Tier 5 (Evaluation) — calibration, class imbalance.
  • Tier 7 (Production Engineering) — all 5 posts.
  • Posts 95 (Anomaly), 109 (Word2Vec), 110 (CV before ViTs for OCR).
  • Tier 6 (Deep Learning) — Transformer, BERT for NER.
  • Compensation

    Cleartax Senior MLE / DS in 2026 ranges roughly ₹28 lakh – ₹50 lakh for 5-7 YOE. Lead roles ₹65 lakh+.

    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 →