Ankit Sharma profile photo 1 on Tal

Ankit Sharma

Senior Data Analyst @ American Express | PySpark, Revenue Forecasting, Python

Delhi

8+ years of work experience

American Express
Jul 20224 yrs 2 mosPresent

Gurugram, Haryana, India

American Express
Jul 20224 yrsJun 2026

Gurugram, Haryana, India

• Architected data science forecasting suite using Python, NumPy, SciPy, Pandas, and deep learning models (XGBoost, ARIMA, SARIMA) on GCP, delivering 95% prediction accuracy across 12+ product lines. • Engineered end to end big data pipeline on GCP leveraging PySpark, Hive, and Python to orchestrate lead allocation for 1.5M daily records, reducing latency by 40%. • Refactored legacy SAS workflows into Python based ETL pipelines with Pandas and NumPy, cutting processing time by 60% and boosting data accuracy for quarterly risk reports. • Directed GenAI Signing Review initiative, applying NLP techniques with NLTK and LLMs (deep learning) to automate risk assessment, cutting manual review time by 30%. • Implemented AI driven risk scoring engine using machine learning, Pandas, and LLM generated insights, raising decision accuracy by 12% and slashing manual reviews by 30%. • Processed >100M transaction records with Python, Pandas, NumPy, and Hive on GCP Cloud Storage, enabling scalable audit reporting and 24 hour data refresh cycles. • Coordinated stakeholder management across data science, engineering, and product groups to deploy GenAI solution, ensuring compliance and on time delivery.
The Smart Cube
Sep 202111 mosJul 2022

Noida, Uttar Pradesh, India

• Built price elasticity and churn prediction models in Python using Pandas, NumPy, and scikit learn regression/classification, influencing pricing strategy for a $200M portfolio. • Designed market mix models with Python and statistical analysis, boosting campaign ROI by 20% across 15 client accounts. • Engineered ETL pipelines using Python, SQL, and Hive to ingest 500K+ records weekly, supporting downstream analytics and reporting.
GlobalLogic
Jul 20183 yrs 3 mosSep 2021

• Led design of fraud detection system employing machine learning and deep learning models in Python, detecting anomalies in 2M+ health claims with 98% precision. • Architected end to end ML lifecycle on a Hadoop cluster, integrating data preprocessing with Pandas and NumPy, model training, deployment, and continuous monitoring. • Optimized ETL pipeline performance using Python, Pandas, and NumPy, reducing runtime by 35% and visualizing data flows with Matplotlib. • Delivered scalable distributed solutions for high volume data processing.
KPMG Global Services (KGS)
Jul 20174 mosOct 2017

Gurgaon, Haryana, India

1. Migrating KPMG Portal pages First we create a webpages Page using scripting language and Adding KPMG inbuild Web parts. 2. KPMG web parts only work with this page layout. KPMG web parts should only be used in predefined zones. 3. Navigation – This feature should be limited to a maximum of three levels of concise, meaningful labels, relevant to a given member firm or function site collection. 4. Web Parts - 11 KPMG Web Parts have been deployed as part of the 5. SharePoint 2013 Portals project. 6. Local Search Refiner - Refiners to be used for generating pre-filtered URLs should be configured and available in the Production environment. 7. Quality Analysis of Pages.

Ankit’s AI Stack

cursor
cursor
opencode
opencode
claudecode
claudecode

Education

Postgraduate Degree, Mobile Computing

CDAC Bangalore

20162016
Uttarakhand Technical University

Engineer's Degree, Information Technology

Uttarakhand Technical University

20114 yrs2015

Join Ankit and 10k+ other professionals on

Tal Boss

Tal is the app where bosses meet engineers & builders directly, without middlemen

Download Tal Boss

Ankit Sharma on Tal

Headline

Senior Data Analyst @ American Express | PySpark, Revenue Forecasting, Python

About

Graduated with an Engineer's Degree in Information Technology from Uttarakhand Technical University and a postgraduate degree in Mobile Computing from CDAC Bangalore. Currently working as a Senior Data Analyst at American Express, contributing to data-driven projects by leveraging PySpark, Python, and advanced analytics to support organizational objectives. Holds multiple certifications, including Google Analytics and Apache PySpark by Example, reflecting a commitment to professional growth. Motivated by opportunities to apply technical expertise in revenue forecasting and data analysis, contributing to impactful business outcomes.

Location

Delhi

AI stack

cursor

opencode

claudecode

Work experience

Analyst - Risk & Info Management at American Express, Jul 2022 - Present · 4 yrs 2 mos, Gurugram, Haryana, India

Senior Data Analyst at American Express, Jul 2022 - Jun 2026 · 4 yrs, Gurugram, Haryana, India, • Architected data science forecasting suite using Python, NumPy, SciPy, Pandas, and deep learning models (XGBoost, ARIMA, SARIMA) on GCP, delivering 95% prediction accuracy across 12+ product lines. • Engineered end to end big data pipeline on GCP leveraging PySpark, Hive, and Python to orchestrate lead allocation for 1.5M daily records, reducing latency by 40%. • Refactored legacy SAS workflows into Python based ETL pipelines with Pandas and NumPy, cutting processing time by 60% and boosting data accuracy for quarterly risk reports. • Directed GenAI Signing Review initiative, applying NLP techniques with NLTK and LLMs (deep learning) to automate risk assessment, cutting manual review time by 30%. • Implemented AI driven risk scoring engine using machine learning, Pandas, and LLM generated insights, raising decision accuracy by 12% and slashing manual reviews by 30%. • Processed >100M transaction records with Python, Pandas, NumPy, and Hive on GCP Cloud Storage, enabling scalable audit reporting and 24 hour data refresh cycles. • Coordinated stakeholder management across data science, engineering, and product groups to deploy GenAI solution, ensuring compliance and on time delivery.

Senior Data Analyst at The Smart Cube, Sep 2021 - Jul 2022 · 11 mos, Noida, Uttar Pradesh, India, • Built price elasticity and churn prediction models in Python using Pandas, NumPy, and scikit learn regression/classification, influencing pricing strategy for a $200M portfolio. • Designed market mix models with Python and statistical analysis, boosting campaign ROI by 20% across 15 client accounts. • Engineered ETL pipelines using Python, SQL, and Hive to ingest 500K+ records weekly, supporting downstream analytics and reporting.

Senior Software Engineer at GlobalLogic, Jul 2018 - Sep 2021 · 3 yrs 3 mos, • Led design of fraud detection system employing machine learning and deep learning models in Python, detecting anomalies in 2M+ health claims with 98% precision. • Architected end to end ML lifecycle on a Hadoop cluster, integrating data preprocessing with Pandas and NumPy, model training, deployment, and continuous monitoring. • Optimized ETL pipeline performance using Python, Pandas, and NumPy, reducing runtime by 35% and visualizing data flows with Matplotlib. • Delivered scalable distributed solutions for high volume data processing.

Intern at KPMG Global Services (KGS), Jul 2017 - Oct 2017 · 4 mos, Gurgaon, Haryana, India, 1. Migrating KPMG Portal pages First we create a webpages Page using scripting language and Adding KPMG inbuild Web parts. 2. KPMG web parts only work with this page layout. KPMG web parts should only be used in predefined zones. 3. Navigation – This feature should be limited to a maximum of three levels of concise, meaningful labels, relevant to a given member firm or function site collection. 4. Web Parts - 11 KPMG Web Parts have been deployed as part of the 5. SharePoint 2013 Portals project. 6. Local Search Refiner - Refiners to be used for generating pre-filtered URLs should be configured and available in the Production environment. 7. Quality Analysis of Pages.

Education

Postgraduate Degree, Mobile Computing, CDAC Bangalore

Engineer's Degree, Information Technology, Uttarakhand Technical University

Ankit Sharma - Analyst - Risk & Info Management in Delhi | Tal