Data Scientist / Machine Learning Engineer (7-9 Years Experience)

Overview

RevOpt is building an AI-native Total Revenue & Profit Optimization System for the hospitality industry. The platform will help hotels move beyond siloed room pricing into coordinated optimization of demand, price, total revenue, and profit, with explainability and governance built in.

We are looking for a hands-on Data Scientist / Machine Learning Engineer (3–5 years) to build the core intelligence layer: forecasting, demand and revenue modelling, pricing support, and decision signals that plug directly into our product. This is a build-and-ship role in a small team, working closely with the founders and our Backend Lead.

What you’ll work on

Design and implement forecasting models for demand, occupancy, ADR, revenue and other key KPIs using real hotel and market data.

Build feature pipelines from PMS feeds, market intelligence, historical bookings, cancellations, and segments to create clean, model-ready datasets.

Develop practical predictive and decision-support models that inform pricing, demand shaping and total revenue / profit optimization, rather than pure research models.

Set up robust evaluation and back-testing frameworks to judge model quality on historic data before pilots.

Collaborate with the Backend Lead to expose model outputs via services or scheduled pipelines that can be consumed by the RMS decision engine.

Produce structured outputs and driver signals (e.g., pickup trends, comp-set shifts) that support explainable recommendations and AI-generated narratives.

Participate in early pilot cycles, analysing results and refining models as we learn from real hotels.

What you bring

3–5 years of applied ML / data science experience in product, SaaS, or analytics teams.

Strong Python skills for data and modelling (pandas, NumPy, scikit-learn; bonus if you know Prophet, XGBoost/LightGBM, or ARIMA-style methods).

Strong SQL and comfort working directly with relational data (PostgreSQL or similar).

Hands-on experience with time-series / forecasting / demand / pricing / optimization problems (at least one serious project).

Ability to take messy operational data, clean and structure it, engineer features, and build models that can be used in production.

A product mindset: you care about commercial relevance, stability and explainability, not just marginal accuracy gains.

Nice to have

Experience with Airflow / Dagster / MLflow or similar tools for ML pipelines and experiment tracking.

Exposure to pricing, revenue management, operations research, or decision-heavy domains (hospitality, travel, fintech, retail).

Experience deploying models behind FastAPI / Flask or working closely with engineers to operationalise models.

Some exposure to LLM-based summarisation or explainability is a plus, not a requirement.

If you want to build a real AI-native product in hospitality, work closely with experienced founders and a strong backend lead, and own the model layer end-to-end, we’d like to hear from you.

Please answer the following when applying:

Describe a forecasting or predictive model you built.
What problem were you solving, what data did you use, and how did you evaluate whether the model was good enough?

Have you worked with time-series, demand, pricing, or operational forecasting data before?
If yes, briefly explain the project and your role.

What is your comfort level with Python, SQL, and deploying or operationalizing ML outputs for use in a product?

Have you worked on messy real-world datasets with missing values, inconsistent mappings, or noisy operational data?
Briefly describe one such situation and how you handled it.

Are you comfortable working closely with backend/product teams in an early-stage startup where models need to be practical, interpretable, and production-usable?