Machine Learning Data Engineer

Kalibri Labs • Washington, DC

1d • Remote

About The Position

At Kalibri, we are helping to redefine and rebuild the hotel industry. We are looking for passionate, energetic, and hardworking people with an entrepreneurial spirit, who dream big and challenge the status quo. We are working on cutting-edge solutions for the industry: we harness cloud-native data pipelines with advanced AI/ML models to drive asset performance. Kalibri is growing, so if you’re ready to make a difference and utilize your talents across a groundbreaking organization, please keep reading!
About the Role
We're looking for a Machine Learning Data Engineer who can build, maintain, and improve the production pipelines that power Kalibri's core algorithmic products — including Census, Prediction, Estimation, and OBM. This role is ideal for someone mid-level who thrives turning complex models into reliable, scalable production systems.
This role will design, build, and maintain production data pipelines. You'll collaborate with Data Science, ML Engineering, Data Operations, and Product on transforming raw hospitality data into production-grade ML features and outputs.
This is a great opportunity to engineer the data backbone of AI-powered hospitality, working with big data and machine learning to help increase asset values.

Requirements

• Master's degree or PhD in Computer Science, Data Science, Statistics, Mathematics, or a related quantitative field (or Bachelor's degree with equivalent experience).
• 3–5 years of professional experience as an ML Engineer, Quantitative Engineer, or Research Scientist.
• Strong proficiency in Python for data pipeline development, scripting, and automation.
• Deep experience with SQL and cloud data warehouses, particularly Snowflake (stored procedures, UDFs, semi-structured data, performance tuning).
• Hands-on experience with workflow orchestration tools such as Prefect, Airflow, or similar (e.g., Dagster, Luigi).
• Proficiency with dbt (dbt Core or dbt Cloud) for SQL-based data transformation and testing.
• Experience working with PySpark or similar distributed computing frameworks for large-scale data processing.
• Strong understanding of data modeling, ETL/ELT patterns, and data warehouse design principles.
• Proficiency with Git version control and collaborative development workflows (Bitbucket preferred).
• Demonstrated ability to operationalize ML models — not just train them — including feature pipelines, model serving, and output validation.
• Excellent cross-functional collaboration skills with proven ability to work alongside data scientists, analysts, and product managers.

Nice To Haves

• Experience with ML frameworks such as CatBoost, LightGBM, Prophet, scikit-learn, or statsmodels — particularly in production pipeline contexts.
• Exposure to linear programming or optimization solvers (PuLP, OR-Tools, Gurobi).
• Experience with Jenkins for CI/CD, job scheduling, and deployment automation.
• Experience with AWS S3 for cloud data storage and file-based data exchange.
• Familiarity with Docker for containerized pipeline execution.
• Experience with entity mapping.
• Knowledge of geospatial data processing (Haversine distance, coordinate systems, mapping APIs such as Google Places and Mapbox).
• Experience with data quality and observability frameworks (Monte Carlo, Great Expectations, or similar).
• Familiarity with Parquet, Arrow, or other columnar data formats.
• Experience with dynamic time warping (dtaidistance), multiprocessing, or advanced time-series techniques.
• Background in hospitality, travel, or other data-rich verticals — including familiarity with metrics such as ADR, occupancy, RevPAR, and COPE.

Responsibilities

• Design, build, and maintain production data pipelines using Python, Prefect, Airflow, Jenkins or any other orchestration framework multi-phase algorithmic workflows.
• Build and optimize advanced SQL transformations in Snowflake, including window functions, CTEs, stored procedures, UDFs, and semi-structured data processing.
• Build and maintain dbt models for data transformation, identity resolution, and slowly changing dimension (SCD Type 2) tracking across 80+ models and multiple pipeline stages.
• Build and maintain feature engineering pipelines that feed ML models including CatBoost gradient boosting, Prophet time-series decomposition, LightGBM regression, and PuLP linear programming solvers.
• Operationalize ML model outputs by integrating predicted ADRs, occupancy forecasts, and optimization results into downstream production tables and Parquet file outputs.
• Integrate and reconcile data from multiple heterogeneous sources including hotel property management systems, rate shop providers, mapping APIs, and market forecast data.
• Work with PySpark for large-scale daily distribution processing, managing partitioning strategies, memory tuning, and efficient Parquet I/O across millions of records.
• Implement and monitor data quality frameworks such as DBT and Monte Carlo.
• Manage CI/CD pipelines using Bitbucket Pipelines for automated testing, linting (SQLFluff), and deployment of dbt projects and Python applications.
• Containerize pipeline components with Docker for consistent execution across development and production environments.
• Implement robust retry logic, error handling, and fallback strategies across pipeline phases to ensure reliable daily and monthly production runs.

Benefits

• Fully remote work, with a thriving company culture
• Robust medical, dental, and vision plans through Blue Cross Blue Shield, including a $0 cost plan for employees and subsidized coverage for dependents
• 401k plan with employer match
• Flexible Paid Time Off
• $250 new hire allowance for home office setup

What This Job Offers

• Job Type: Full-time
• Career Level: Mid Level
• Number of Employees: 11-50 employees