Data Scientist-1

Location: Athens, GR

Remote Type: Hybrid

Time Type: Full time

Job Description

Job Description:

Strategic Imperative: 

The Data Scientist role with a focus on Predictive Modeling, is responsible for developing, evaluating, and refining analytical and machine learning models that predict user behavior and optimize business outcomes across our suite of products. This role focuses on modeling, feature development, and offline evaluation to support performance marketing, insights  and other strategic initiatives.

The position applies statistical and machine learning techniques to large-scale behavioral and transactional data to improve ranking, recommendation and yield optimization. While this role partners closely with Engineering and ML Engineering teams, it does not own production deployment, infrastructure, or MLOps.

Who We Are!  

Pollfish, a Prodege, LLC company, is an online market research survey platform where data driven brands bring market research in-house for faster and smarter decision making. We have a proprietary network of 250M consumers/year which enables companies to connect with and understand real consumers worldwide in a fast, easy and cost-effective way.

Primary Objectives: 

• Improve Yield Through Predictive Modeling: Drive yield optimization through the ideation and development of machine learning models for recommendations, ranking and yield optimization use cases

• Organizational Data Science Advancement: Expand the application of machine learning across business functions through identification of opportunities and modeling, analysis, and offline evaluation.

• Data Design: Leading feature engineering and defining key concepts to be formalized within the feature store.

Qualifications - To perform this job successfully, an individual must be able to perform each job duty satisfactorily. The requirements listed below are representative of the knowledge, skill, and/or ability required. Reasonable accommodations may be made to enable individuals with disabilities to perform the essential functions.

Detailed Job Duties: (typical monthly, weekly, daily tasks which support the primary objectives)

• Business Translation & Modeling Framework Design:

  • Partner with Product and other business stakeholders to frame business problems into well-defined modeling objectives and hypotheses.

  • Select appropriate modeling techniques (e.g., classification, regression, deep learning, reinforcement learning) based on use case and data characteristics.

  • Define success metrics and evaluation criteria that align model performance with business impact.

  • Document modeling assumptions, tradeoffs, and intended use to ensure clarity, transparency, and alignment across teams

• Data Analysis & Feature Development

  • Perform analysis on large-scale behavioral and transactional datasets to identify patterns, drivers, and potential predictive signal

  • Engineer features from user behavior, lifecycle activity, offer attributes, and revenue data to support modeling objectives

  • Collaborate with ML team to ensure features properly registered within feature store

• Model Development & ML Collaboration:

  • Leverage appropriate modeling frameworks and packages to build high performing predictive models

  • Perform model tuning, validation, and comparison using appropriate metrics, cross-validation, and offline testing frameworks

  • Interpret model outputs to assess business relevance, identify strengths and limitations, and validate against observed behavior

  • Leverage appropriate frequentist and Bayesian approaches to measure model performance and define strategies for balancing exploration and exploitation

  • Document modeling approaches, performance results, and learnings in model cards to support reuse, iteration, and long-term knowledge building

  • Enable ML team to deploy models; collaborate on retraining and maintenance plans

What does SUCCESS look like?

Success in this role is demonstrated by the delivery of predictive models that materially improve yield management and optimizes engagement and revenue. Over time, success is reflected in models that are trusted by stakeholders, features that consistently capture meaningful behavioral signals, and clear evidence that model-driven decisions outperform prior approaches. Strong collaboration with Analytics Engineering and Machine Learning partners, thorough documentation of methods and learnings, and measurable business impact are hallmarks of high performance in this role.

The MUST Haves: (ex: job cannot be done without these skills, education, experience, certifications, licenses) 

• Bachelor’s or Master’s degree in Computer Science, Data Science, Statistics, Mathematics, or a related quantitative field

• Three or more (3+) years experience performing deep data analysis and  training machine learning models

• Strong foundation in machine learning and statistical modeling (classification, regression, ranking, optimization)

• Experience with ranking, recommendation systems, and personalization models

• Fluency in Python and common ML libraries (scikit-learn, XGBoost/LightGBM, PyTorch, or TensorFlow)

• Fluency in SQL and ability to work with large, event-level datasets in data warehouse environments (e.g., Snowflake, BigQuery, Redshift)

• Experience with feature engineering, model evaluation, and performance diagnostics

• Strong analytical reasoning and ability to translate business questions into modeling approaches

• Clear communication skills, particularly in explaining model results and tradeoffs to non-technical stakeholders

• Understanding of ML Ops concepts and the ability to collaborate effectively with ML Engineering and ML Ops teams 

• Excellent attention to detail

• Proficiency in critical thinking and problem solving.

The Nice to Haves: (preferred additional skills, education, experience, certifications, licenses) 

• Hands-on experience managing ML deployments and designing feature stores and registries

• Experience with AI/ML frameworks such as LangChain, LLMs, and HuggingFace

• Worked with distributed data processing frameworks (Spark, Ray, Flink, Trino).

• Experience with ML experiment tracking (e.g., ML Flow)

• Experience in loyalty programs or performance marketing or market research

• Experience with contextual multi armed bandit algorithms or reinforcement learning

• Preference modeling methods used in Conjoint/MaxDiff

• Understanding of Bayesian statistics inference (e.g., PyMC)

Perks & Benefits:

• An attractive salary package

• Part of an innovative Global Tech Company

• Private Health Insurance

• Company Equity

• Weekly Office Events - Catered Lunch and Breakfast

• Stocked Kitchen

• Company Outings & Quarterly Events

• Hybrid Working

• Meal Coupons - Monthly

• LinkedIn Learning & Training Opportunities/Budget

• Mental Health Benefits - Wellness Coach App Subscription

• Great office location in the city center - Parking slots available

• Gym Subscription - UP Fit

• Quarterly Charitable Giving Allowance

• Peer recognition Allowance