About Deepika
🇫🇷 version :
- RAG pipelines & LLM integration
- ETL automation & data quality
- Analytics dashboards ready for stakeholders
- A/B testing frameworks end-to-end
🇬🇧 version :
- RAG pipelines & LLM integration
- ETL automation & data quality
- Analytics dashboards ready for stakeholders
- A/B testing frameworks end-to-end
English
Native or bilingual
French
Conversational
Tamil
Native or bilingual
Experience
- Freelance (ChargeMOD, Vydyuthi Energy services, EURECOM)Data Analyst and Data ScientistMarch 2024 - Today (2 years and 3 months)
- Delivered end-to-end product analytics reports to optimize sales funnels — improving conversion rate, average order value, and time to convert
- Drove the full A/B testing cycle — hypothesis formulation, methodology, monitoring, and results analysis — with direct impact on conversion, upsell, and cross-sell
- Designed and deployed data models across several use cases: marketing attribution, app/web usage, churn, upsell, and cross-sell — with stakeholder-ready dashboards
- Automated ETL pipelines with reusable Python scripts and built-in quality checks, reducing manual effort
- Trained business teams on dataviz tools and created methodological guides and A/B testing frameworks to ensure analysis reliability and reproducibility
- JK
On Malt
Data AnalystDIGITAL AND ITDecember 2025 - January 2026 (1 month)- Conducted in-depth product data analysis to identify customer journey friction points and optimize the conversion funnel — tracking conversion rate, average order value, and time to convert
- Designed and modeled data structures suited for marketing and business needs, developed a strategic analytics roadmap, and deployed tracking models and interactive dashboards for operational teams
- Supported business teams with dataviz tools and created methodological guides and A/B testing frameworks for reliable and reproducible analyses
- STMicroelectronicsData Science InternMay 2023 - November 2023 (6 months)France
- Designed a CNN gesture classifier from scratch using ToF sensor data — achieving 96% accuracy through Bayesian hyperparameter optimization with Optuna
- Developed an incremental learning system enabling the model to incorporate new gestures without forgetting existing knowledge — directly addressing the catastrophic forgetting problem
- Collected and cleaned raw gesture recognition data from a Time-of-Flight sensor, building a reliable pipeline ready for modeling
- Applied L1, L2, and Elastic Weight Consolidation (EWC) regularization techniques to maintain stable performance on evolving datasets
Reviews
Recommendations
These freelancer profiles also match your criteria
Agatha Frydrych
Backend Java Software Engineer
4.7
(3)
2
Baptiste Duhen
Fullstack developer
4.6
(4)
5
Amed Hamou
Senior Lead Developer
4
(2)
7
Audrey Champion
Web developer
4.3
(3)
4
Education
- Masters in Data Science and EngineeringEURECOM2024Cours suivis : Apprentissage automatique, Apprentissage profond, Statistiques, Systèmes de gestion de bases de données, Simulation d’entreprise, Sécurité des systèmes, Information quantique.
- Bachelors in Computer Science and EngineeringSSN College of Engineering2021Cours suivis : Apprentissage automatique, Développement web, Développement Android, Cryptographie, Systèmes d’exploitation, Conception et analyse d’algorithmes, Conception de compilateurs.