Data Scientist (Platform Modernization – Production ML)

CONA Services
Atlanta, US
Hybrid

Job Description

We are seeking a highly motivated Data Scientist to help modernize an existing production analytics platform comprised of two applications: (1) recommendations for convenience retail and FSOP, and (2) product availability. In this role, you will enhance and operationalize ML/AI solutions end-to-end—partnering with product, engineering, and business stakeholders to improve model performance, reliability, scalability, and time-to-value. You will also contribute to data engineering efforts where necessary to ensure trusted, well-modeled, and production-ready data pipelines that power these applications.

Responsibilities

Platform Modernization & Applied Data Science (Primary)

  • Partner with product, engineering, and business stakeholders to modernize and scale a production platform supporting recommendations (convenience retail & FSOP) and product availability.
  • Assess current models, features, and data flows; prioritize technical debt and propose a pragmatic modernization roadmap (accuracy, latency, robustness, maintainability).
  • Build, validate, and deploy ML/analytics solutions using production-grade patterns (reproducible training, versioning, automated testing).
  • Establish measurement and experimentation loops (offline evaluation, online testing where applicable) and quantify impact of increments released.
  • Communicate tradeoffs, results, and recommendations through clear narratives and visualizations for technical and non-technical audiences.
  • Operational excellence: define and monitor model/application health (data quality checks, drift detection, performance SLAs) and drive continuous improvement in partnership with platform/architecture teams.

Data Engineering – Secondary Focus

  • Partner on scalable ingestion and transformation pipelines (e.g., Azure Databricks, Azure Data Factory) that support both recommendation and availability use cases.
  • Implement and maintain reliable feature and training datasets, including data validation and lineage to support production ML.
  • Contribute to lakehouse patterns for batch and near-real-time processing; collaborate with teams using event-streaming technologies where applicable.
  • Support integration patterns (APIs, jobs, and services) required to operationalize models and analytics into the two platform applications.

What will you learn?

  • Deep understanding of bottler operations and industry-specific analytics applications.
  • Data science, Machine learning, and broader AI are highly impactful to achieve meaningful business outcomes. You will get to apply your skills to real-life business problems.
  • Industry/FMCG trends and Benchmarks for new/emerging technologies incl. vendor roadmaps and strategic developments.
  • Bottler and NAOU (North American Operating Unit, Coca-Cola Company) business strategies.

What makes you a good fit?

Minimum Qualifications

  • Bachelor's degree in computer science, Statistics, Mathematics, or a related field (or equivalent practical experience).
  • 4+ years in data science (or closely related applied ML/analytics role), delivering end-to-end solutions in production environments.
  • Hands-on expertise building and evaluating machine learning models (e.g., scikit-learn, XGBoost, LightGBM, time-series and/or deep learning architectures).
  • Proficiency in Python and SQL.
  • Experience deploying and operating models in production, including monitoring, performance measurement, and iteration based on feedback.
  • Ability to work within an existing platform/codebase, identify modernization opportunities, and deliver improvements incrementally without disrupting service.

Preferred Qualifications

  • Experience partnering with data engineering and/or MLOps teams (or owning parts of DE/MLOps work) to productionalize ML systems (CI/CD, automated testing, release practices).
  • Experience building reliable ETL/ELT pipelines and working with structured and unstructured data.
  • Proficiency with Databricks, PySpark, Azure Data Factory, and Azure Data Lake (or comparable cloud tooling).
  • Familiarity with common ML operations patterns (feature/training data management, lineage, reproducibility, monitoring).
  • Generative AI familiarity (e.g., using LLM tools to accelerate development, improve explainability, or support analysis of workflows).

Professional & Interpersonal Skills

  • Strong analytical thinker with proven problem-solving abilities.
  • Exceptional written, verbal, and interpersonal communication skills.
  • Adaptable; thrives in fast-paced, dynamic environments with shifting priorities.
  • Collaborative team player with the ability to influence stakeholders across functions.
  • Committed to fostering diversity, equity, and inclusion in the workplace.
  • Consistently demonstrates CONA’s core values: Integrity, Accountability, Passion, Collaboration, and Innovation

Work Environment: CONA follows a hybrid work model requiring a minimum of 3 days (60%) in the office per week to support collaboration and development. Tuesdays and Wednesdays are

Skills & Requirements

Technical Skills

Machine learningData sciencePythonSqlScikit-learnXgboostLightgbmTime-seriesDeep learningAzure databricksAzure data factoryEvent-streaming technologiesCommunicationCollaborationPlatform modernizationProduction mlData engineeringRecommendationsProduct availabilityBottler operationsIndustry analytics

Employment Type

FULL TIME

Level

Mid-Level

Posted

4/28/2026

Continue to LinkedIn

You will be redirected to the job posting on LinkedIn.

Sign in and we'll score your resume against this role.