Aman Jethani

Data Scientist

Banda, India3 yrs 7 mos experience

Most Likely To SwitchAI ML Practitioner

Key Highlights

Reduced RMSE by ~23% in model performance.
Developed a fraud detection model with a 10% GINI lift.
Created automated workflows improving model retraining efficiency.

Stackforce AI infers this person is a Data Science professional with a focus on Fintech and EdTech industries.

Contact

Skills

Core Skills

Machine LearningData ScienceData EngineeringNatural Language ProcessingData Analysis

Other Skills

Amazon Web Services (AWS)Apache SupersetAutomationAutoregressive ModelsBERT (Language Model)Back-End Web DevelopmentC (Programming Language)C++CatBoostData ManipulationData Pipeline DevelopmentDeep LearningElasticsearchExploratory Data AnalysisFeature Engineering

Experience

3 yrs 7 mos

Total Experience

1 yr 2 mos

Average Tenure

2 yrs 9 mos

Current Experience

Quantumstreet ai

2 roles

Data Scientist

Promoted

Jan 2025 – Present · 1 yr 5 mos

1) Enhanced the Expected Return (ER) model through advanced feature engineering, reducing RMSE by ~23% and outperforming the WatsonML platform in ~70% of use cases.
2) Designed and implemented a custom loss function in CatBoost to improve directional accuracy, achieving a ~27% boost in prediction directionality.
3) Developed an auto-scaling infrastructure and automated pipeline for data processing, model training, and inference, resulting in a ~30-40% reduction in training time.
4) Built an automated retraining workflow for degraded models, leading to a ~5-7% increase in overall model performance.
5) Consistently outperformed SPY benchmarks, demonstrating the robustness and effectiveness of the enhanced ER model in live trading environments.

Exploratory Data AnalysisTerraformAmazon Web Services (AWS)Autoregressive ModelsElasticsearchIBM Cloud+13

Associate Data Scientist

Aug 2023 – Dec 2024 · 1 yr 4 mos

Aplazo

Data Science Intern

Jan 2023 – Aug 2023 · 7 mos

1) Developed transaction level fraud detection model for offline and online BNPL new customers giving a ~10% lift in GINI compared to previous versions.
2) Reduced overall delinquency rate (DOB22) in the month of May 2022 which leads less fraudulent customers.
3) Conducted multiple root cause analyses on data acquisition strategies to optimize the onboarding process, resulting in improved efficiency and reduced risk.

Exploratory Data AnalysisSQLPython (Programming Language)Machine LearningData Science