Data Scientist

Role: Data Scientist

Years of Experience: 2-4 years

Job Location: Bangalore (Indiranagar)

Work Type: Hybrid

Shift Timing : 9:30 AM to 6:30 PM

Notice Period : Looking for candidate who can join us in 30-60 days

Responsibilities:

We are seeking a specialized Data & ML Engineer with 4 Yr Experience to implement our advanced analytics strategy within the Amazon Marketing Cloud (AMC) ecosystem. You will be successful in this role if you are used to leveraging AMC's clean room environment to build sophisticated machine learning models that drive high-value marketing outcomes, with experience building predictive audiences and attribution models.

You will have skills beyond standard SQL, reporting and implement high-impact ML workflows within AMC, including:

Custom Model Implementation: Build and deploy proprietary modeling algorithms using AMC Custom Models for Audiences. You will create high-signal segments for Customer Lifetime Value (CLV) and New-to-Brand (NTB) prediction

AWS Clean Rooms ML & Lookalike Modeling: Utilize AWS Clean Rooms ML to train lookalike models on seed audiences (e.g., top 10% of high-value customers)

Multi-Touch Attribution (MTA) & Sequencing: Develop ML-driven attribution models to understand the optimal sequence of ad exposures

Predictive Audience Discovery: Use ML to identify value-seeking shoppers and high-propensity segments by joining GCP-hosted CRM data with Amazon's pseudonymized signals

Media Optimization & Frequency Analysis: Build models to identify Ad Saturation Pointspredicting the exact frequency cap where ad effectiveness diminishes to optimize ROAS

You will have proven experience in:

Ingesting data it into AMC (S3/AWS) for join-based ML modeling

Using Python and SQL to automate data refreshes

Ingestion of model-based scores into media activation platforms

Managing the lifecycle of models within a privacy-safe environment

Hashing data for privacy compliance

Ensuring model drift is monitored and training datasets are updated (using ETL services like AWS Glue and Dataflow).

Designing pipelines to export first-party data from GCP (BigQuery)

You will be successful in this role if you have the following skills:

Deep hands-on experience with Amazon Marketing Cloud (AMC) and AWS Clean Rooms.

Experience with BigQuery, Vertex AI, and Cloud Storage as the primary source of truth for first-party data.

Strong command of Python (Scikit-learn, Pandas) for feature engineering and model development.

Expertise writing complex JOINs and aggregations within the unique privacy constraints of AMC (e.g., handling aggregation thresholds).

Domain knowledge of Amazon DSP, Sponsored Ads, and the mechanics of retail media.

You don't need to have all the skills to be suitable for the role. An outstanding candidate will also have experience with the following:

Differential Privacy, K-anonymity, and Hashing (SHA-256) protocols.

Ability to work with the Amazon Ads API for automated audience pushes and reporting.

Ability to translate Model Lift and Probability Scores into actionable budget shifts for Marketing stakeholders.