Eliorexa
ATS versionView 3D portfolio

Data Scientist · Mid-level · ~5 yrs

Karthik Nair

Senior Data Scientist

Bengaluru, India karthik.nair@hey.com +91 99014 26839 linkedin.com/in/karthik-nair-ds github.com/karthiknair
View interactive 3D portfolio
ATS / printable PDF

Professional summary

Data scientist with 5 years building models and experiments that move business metrics in e-commerce and fintech. Owns problems end to end — framing the question with stakeholders, engineering features in Python/SQL, training and shipping machine-learning models, and proving impact through A/B testing. Translates statistics into plain-language recommendations and mentors junior analysts on rigour and reproducibility.

4M

Monthly shoppers modelled

12%

Lift in add-to-cart rate

₹1.4Cr

Annual overstock saved

Skills

Languages

PythonRSQLPySpark

ML & experimentation

scikit-learnXGBoostmachine learningA/B testingcausal inferencestatisticsforecasting

Data & MLOps

pandasNumPydbtAirflowMLflowDockerAWS SageMaker

Visualization & BI

TableauLookermatplotlibseabornStreamlit

Work experience

Data Scientist · Kettl Commerce (Series C marketplace)

Aug 2022 – Present

Bengaluru

  • Own the recommendations and ranking models (Python, XGBoost) serving 4M monthly shoppers; an experiment-driven re-ranker lifted add-to-cart rate 12% and revenue per session ₹38.
  • Built the company A/B testing framework with guardrail metrics and sequential testing, cutting average experiment runtime 30% while controlling false-positive rate.
  • Developed a demand-forecasting model that reduced overstock write-offs ~₹1.4 crore/year across the top 200 SKUs.
  • Mentored 2 junior analysts and ran a monthly statistics clinic, standardising how the team reports significance and confidence intervals to stakeholders.

Data Analyst → Junior Data Scientist · Paysprout

Jun 2020 – Jul 2022

Remote

  • Created a credit-risk scorecard in Python/scikit-learn that reduced default rate 17% while approving 8% more applicants.
  • Automated SQL + pandas reporting pipelines feeding Tableau, replacing manual decks and saving ~10 analyst-hours/week.
  • Designed and analysed 25+ A/B tests on the checkout funnel, with two winning variants adding an estimated ₹62L annual revenue.

Featured projects

3 teams
Liftlab — open-source experiment analysis

A Python toolkit for A/B test power analysis and CUPED variance reduction, adopted across 3 internal teams.

  • Implemented CUPED and sequential testing, shrinking required sample sizes ~25% and shortening time-to-decision.
PythonA/B testingstatistics
Interactive 3D Portfolio

WebGL portfolio built on Eliorexa, linked from this résumé.

  • Reactive three.js hero with scroll-driven, reduced-motion-safe sections showcasing models and dashboards — LCP 1.1s on mobile.
three.jsReactdata visualization

Education

M.Sc. Applied Statistics

2020

Christ University, Bengaluru

First Class with Distinction

Certifications

AWS Certified Machine Learning – Specialty

Amazon Web Services · 2023

See the work in 3D

Explore Karthik's interactive WebGL portfolio — projects, skills and a way to get in touch.

Open the 3D portfolio