Data Scientist

Role :- Data Scientist

Experience :- 3 to 5 Years

Contract Tenure :- 6 Months

Location :- Remote

Job Description: Recommendation Engineer (Machine Learning)

About the role

We're looking for a Recommendation Engineer to design, build, and optimize machine

learning systems that personalize user experiences across content, products, or services.

You'll work end-to-endfrom data and feature engineering to model training, deployment,

and experimentationpartnering closely with Product, Data, and Platform teams.

What you'll do (Key Responsibilities)

Design and develop recommendation models using collaborative filtering, content-

based methods, and hybridapproaches (including ranking and retrieval

architectures).

Build and maintain scalable pipelines for user behavior collection, feature

engineering, and batch + near-real-time data processing (e.g., Spark, SQL,

streaming).

Train, evaluate, and fine-tune models using modern ML frameworks; productionize

training and inference workflows on cloud infrastructure (e.g., managed training,

endpoints, CI/CD for ML).

Define success metrics (CTR, engagement, retention, conversion), run A/B tests,

and iterate based on experiment outcomes and model monitoring.

Implement model observability: data drift, model drift, bias/fairness checks, and

automated retraining/rollback strategies.

Collaborate with engineering teams to integrate recommendation services into APIs,

apps, and data products with reliability, latency, and scalability constraints.

Document system design, trade-offs, and model behavior for technical and non-

technical stakeholders.

What we're looking for (Required Qualifications)

3+ years of experience building and shipping ML models in production, ideally

personalization or recommender systems.

Strong foundations in machine learning, statistics, and optimization, with practical

experience in ranking/recall problems.

Proficiency in Python and strong SQL skills; experience with distributed data

processing (e.g., Spark).

Experience deploying models to production (batch or real-time), including model

packaging, inference APIs, and monitoring.

Solid understanding of experimentation and causal measurement (A/B testing,

guardrail metrics, statistical significance).

Ability to communicate clearly, write design docs, and work cross-functionally.

Preferred Qualifications (Nice to Have)

Experience with large-scale personalization: embeddings, candidate generation +

ranking stacks, multi-stage recommenders.

Familiarity with managed ML platforms such as Amazon SageMaker on Amazon

Web Services (training jobs, pipelines, feature stores, endpoints).

Knowledge of recommender-specific techniques: matrix factorization, implicit

feedback modeling, two-tower models, sequential/session-based recommenders.

Experience with streaming systems (Kafka/Kinesis), feature stores, and online/offline

feature parity.

Familiarity with cloud ecosystems like Google Cloud and modern MLOps tooling

(model registry, experiment tracking).

Exposure to privacy-safe personalization, PII handling, and responsible AI practices.

Tech stack

Languages: Python, SQL

Data: Spark, data warehouses/lakes, streaming pipelines

ML: common deep learning frameworks, embeddings, ranking models

MLOps: CI/CD, model registry, monitoring/alerting, experiment tracking

Cloud: managed training + inference services, containerization