Kaustabh Ganguly profile photo 1 on Tal

Kaustabh Ganguly

Senior AI/ML engineer @ Nielsen | M.Tech - AI @ IIT Madras

Greater Bengaluru Area

8+ years of work experience

IEEE SMC KGEC
May 20264 mosPresent

Mentoring Students for IEEE SMC KGEC chapter.
Gracenote, Nielsen Company
Dec 20259 mosPresent

Greater Bengaluru Area

GitHub
Jan 20188 yrs 8 mosPresent

Remote

Johnson & Johnson MedTech
Sep 20232 yrs 4 mosDec 2025

• Architected CCP data platform end-to-end (ingestion → validation → feature pipelines) with SOLID design and near-zero rework. • Built ODR (On-Demand Refresh): one-click ad-hoc ETL → 0 orchestrator failures over 8+ months; 0 refresh tickets; faster post-fix re-triggers. • Built FPM (Fail-Proof Mechanism): Excel→Parquet + dynamic Redshift DDL → incidents reduced from 15+ to ~2 per month (~87%↓); ~80% fewer support tickets. • Shipped 4 production ML models across distinct business questions: Analyst-action recommendation (classification) with SHAP waterfall explanations to analysts. • Prototyped natural-language Q&A over Tableau (LangChain) to speed self-serve analytics. • Established MLOps practices: Redshift feature stores, experiment tracking, approvals, post-deploy monitoring, BA feedback loops; internal Flask APIs; CI/CD on Jenkins/Bitbucket. Stack: Python, scikit-learn/XGBoost, SHAP, SQL (Redshift/Postgres), Flask, Tableau, LangChain, AWS (S3/EC2/RDS/Redshift), Jenkins/Bitbucket, MLOps.
Johnson & Johnson MedTech
Sep 20221 yrAug 2023

• Productionized “Data Sentry” for onboarding/offboarding (DB access + AD groups) across AWS/Azure/GCP on Lambda + S3 with OAuth/Microsoft Graph. • Delivered 10k+ lines of Redshift SQL in ~3 months → zero defects to production. • Enhanced self-service access for platform teams, significantly reducing operational toil and expediting SQL delivery. Stack: Python/Pandas, AWS (Lambda/S3), OAuth/Microsoft Graph, Redshift SQL, Git/Jenkins.
Johnson & Johnson MedTech
Oct 20201 yr 11 mosAug 2022

• Ran production ETL on AWS (Redshift/RDS/EC2/S3) orchestrated with Control-M; authored complex SQL for analytics and reporting. • Built S3-based cross-database transfers and dirty-data filters → eliminated delimiter-related ETL failures. • Tuned legacy Redshift views → ~3× faster runtimes and lower compute cost. • Drove 50+ production releases with <2% rework; SDET on 20+ SIT cycles with 0% rework. • Mentored peers on SQL patterns, release hygiene, and on-call runbooks. Stack: Python, SQL/Redshift, AWS (S3/EC2/RDS), Control-M, Jenkins, Bitbucket/Git.
Param.ai
Jun 20182 mosJul 2018

Hyderabad Area, India

I played a key role in advancing Param.ai's hiring automation efforts through the development of a sophisticated similarity matching algorithm. • Leveraged advanced AI techniques to enhance candidate-job fit, improving recruitment efficiency. • Utilized Jaccard similarity and cosine distances on a custom-trained Word2Vec model to optimize matching processes. • Actively contributed to the startup's vision of transforming the hiring landscape for top candidates.
GitHub
@stabgan

Kaustabh’s AI Stack

antigravity
antigravity
codex
codex
cursor
cursor

Education

Indian Institute of Technology, Madras

Master of Technology - MTech, Artificial Intelligence

Indian Institute of Technology, Madras

20232 yrs2025
Kalyani Government Engineering college

Bachelor of Technology - BTech, Information Technology

Kalyani Government Engineering college

20164 yrs2020

Join Kaustabh and 10k+ other professionals on

Tal Boss

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

Download Tal Boss

Kaustabh Ganguly on Tal

Headline

Senior AI/ML engineer @ Nielsen | M.Tech - AI @ IIT Madras

About

Completed Master of Technology (MTech) in Artificial Intelligence at the prestigious Indian Institute of Technology, Madras, and currently contributing as a Senior Member of Technical Staff at Gracenote, a Nielsen company. Core competencies include data engineering, machine learning, and MLOps practices, with a strong foundation in Python, Redshift, and AWS. Academic and professional experiences are complemented by a Bachelor of Technology in Information Technology from Kalyani Government Engineering College. At Gracenote, collaborates on cutting-edge data-driven solutions, leveraging expertise in data pipelines, machine learning modeling, and scalable architectures to deliver impactful outcomes. Passionate about fostering innovation and driving efficiency, they bring a collaborative approach to solving complex challenges. Dedicated to creating sustainable solutions that align with organizational growth and promote a culture of teamwork and progress.

Location

Greater Bengaluru Area

AI stack

antigravity

codex

cursor

Work experience

Mentor - Artificial Intelligence / Data Science at IEEE SMC KGEC, May 2026 - Present · 4 mos, Mentoring Students for IEEE SMC KGEC chapter.

Senior Member of Technical Staff at Gracenote, Nielsen Company, Dec 2025 - Present · 9 mos, Greater Bengaluru Area

Open Source Contributor at GitHub, Jan 2018 - Present · 8 yrs 8 mos, Remote

Lead Data Scientist at Johnson & Johnson MedTech, Sep 2023 - Dec 2025 · 2 yrs 4 mos, • Architected CCP data platform end-to-end (ingestion → validation → feature pipelines) with SOLID design and near-zero rework. • Built ODR (On-Demand Refresh): one-click ad-hoc ETL → 0 orchestrator failures over 8+ months; 0 refresh tickets; faster post-fix re-triggers. • Built FPM (Fail-Proof Mechanism): Excel→Parquet + dynamic Redshift DDL → incidents reduced from 15+ to ~2 per month (~87%↓); ~80% fewer support tickets. • Shipped 4 production ML models across distinct business questions: Analyst-action recommendation (classification) with SHAP waterfall explanations to analysts. • Prototyped natural-language Q&A over Tableau (LangChain) to speed self-serve analytics. • Established MLOps practices: Redshift feature stores, experiment tracking, approvals, post-deploy monitoring, BA feedback loops; internal Flask APIs; CI/CD on Jenkins/Bitbucket. Stack: Python, scikit-learn/XGBoost, SHAP, SQL (Redshift/Postgres), Flask, Tableau, LangChain, AWS (S3/EC2/RDS/Redshift), Jenkins/Bitbucket, MLOps.

Lead Data Engineer at Johnson & Johnson MedTech, Sep 2022 - Aug 2023 · 1 yr, • Productionized “Data Sentry” for onboarding/offboarding (DB access + AD groups) across AWS/Azure/GCP on Lambda + S3 with OAuth/Microsoft Graph. • Delivered 10k+ lines of Redshift SQL in ~3 months → zero defects to production. • Enhanced self-service access for platform teams, significantly reducing operational toil and expediting SQL delivery. Stack: Python/Pandas, AWS (Lambda/S3), OAuth/Microsoft Graph, Redshift SQL, Git/Jenkins.

Data Engineer at Johnson & Johnson MedTech, Oct 2020 - Aug 2022 · 1 yr 11 mos, • Ran production ETL on AWS (Redshift/RDS/EC2/S3) orchestrated with Control-M; authored complex SQL for analytics and reporting. • Built S3-based cross-database transfers and dirty-data filters → eliminated delimiter-related ETL failures. • Tuned legacy Redshift views → ~3× faster runtimes and lower compute cost. • Drove 50+ production releases with <2% rework; SDET on 20+ SIT cycles with 0% rework. • Mentored peers on SQL patterns, release hygiene, and on-call runbooks. Stack: Python, SQL/Redshift, AWS (S3/EC2/RDS), Control-M, Jenkins, Bitbucket/Git.

Machine Learning Intern at Param.ai, Jun 2018 - Jul 2018 · 2 mos, Hyderabad Area, India, I played a key role in advancing Param.ai's hiring automation efforts through the development of a sophisticated similarity matching algorithm. • Leveraged advanced AI techniques to enhance candidate-job fit, improving recruitment efficiency. • Utilized Jaccard similarity and cosine distances on a custom-trained Word2Vec model to optimize matching processes. • Actively contributed to the startup's vision of transforming the hiring landscape for top candidates.

Education

Master of Technology - MTech, Artificial Intelligence, Indian Institute of Technology, Madras

Bachelor of Technology - BTech, Information Technology, Kalyani Government Engineering college

Public links

github: https://github.com/stabgan