Senior Machine Learning Engineer

Job Description

Role Overview

The Senior Machine Learning Engineer is responsible for designing, developing, deploying, and maintaining production-grade machine learning models that support real estate and facilities management decisions. This role requires end-to-end ownership of ML models, including problem framing, data preparation, model training, validation, deployment, monitoring, and retraining. The engineer will ensure model quality, reliability, and compliance with privacy and safety standards while collaborating closely with cross-functional teams such as Product, Platform Engineering, Data Platform, and domain experts.

Key Responsibilities

• Model Design & Delivery: Design, train, and deploy ML models for forecasting space, occupancy, headcount, rent, and financial metrics. Own models end-to-end, including API versioning, documentation, and portability across cloud platforms using containerization.
• Evaluation, Quality & Trust: Lead model evaluation using backtesting, held-out datasets, and appropriate metrics. Identify and resolve quality issues such as data drift, leakage, or model regressions. Monitor production models for accuracy, drift, degradation, and cost, and manage retraining or rollback as needed. Ensure model safety by handling customer data privacy, PII, and uncertainty disclosure.
• Cross-Functional Collaboration: Partner with Product to implement accuracy thresholds and confidence disclosures. Coordinate API contracts and failure behaviors with AI agent teams. Collaborate with Platform Engineering and Data Platform teams for deployment and data curation. Support domain teams and mentor engineers on ML best practices.

Projects and Contributions

The job description does not specify explicit project names or dates; however, the following key project contexts and responsibilities are identified:

Production ML Models for Real Estate and Facilities Forecasting

• • Role: Lead ML Engineer responsible for end-to-end model lifecycle management.
  • Responsibilities: Framing forecasting problems related to space utilization, occupancy, headcount, rent, and financial planning. Designing and training models, deploying them as versioned APIs, and ensuring portability across cloud platforms.
  • Technologies & Methodologies: Python, SQL, ML frameworks such as scikit-learn and PyTorch, service frameworks like FastAPI and Pydantic, containerization for deployment, CI/CD pipelines, and AI coding assistants.

Model Evaluation and Quality Assurance Framework

• • Role: Owner of model evaluation processes and quality assurance.
  • Responsibilities: Implementing time-aware validation, backtesting against baselines, holdout dataset design, metric selection, calibration, and error analysis. Building regression suites integrated into CI to gate model releases. Diagnosing root causes of quality failures and applying fixes.
  • Technologies & Methodologies: Statistical and probabilistic methods for forecasting, prediction intervals for uncertainty quantification, continuous integration tools, and shared evaluation frameworks provided by the platform team.

Model Serving and MLOps

• • Role: Engineer responsible for reproducible deployment, monitoring, and retraining of models.
  • Responsibilities: Managing experiment tracking, model registries, containerized deployment, batch and online inference, and monitoring for drift and silent degradation. Ensuring compliance with privacy and tenant data access rules.
  • Technologies & Methodologies: Containers, experiment tracking tools, model registries, batch and online inference pipelines, privacy testing methodologies.

Embedding, Retrieval, and Synthetic Data Pipelines

• • Role: Contributor to embedding and retrieval system development and synthetic data generation.
  • Responsibilities: Measuring retrieval quality on real customer documents, developing synthetic data pipelines for safe development and testing, evaluating synthetic data fidelity and privacy leakage risks.
  • Technologies & Methodologies: Embedding techniques, retrieval quality metrics, synthetic data generation methods, privacy attack evaluation including membership inference, differential privacy concepts.

Cross-Functional Collaboration and Mentorship

• • Role: Partner and mentor within cross-functional teams.
  • Responsibilities: Collaborating with Product, AI agent teams, Platform Engineering, and Data Platform teams to define API contracts, accuracy standards, and failure behaviors. Feeding model performance data into analytics warehouses. Supporting domain teams and mentoring engineers on ML practices.
  • Technologies & Methodologies: API design, data warehousing, ML best practices, mentoring and knowledge sharing.

Required Skills and Experience

• Minimum 5 years of professional engineering experience with at least one ML model taken from problem framing to production and maintained through retraining and drift management.
• Proficiency in Python and SQL, including ML libraries (scikit-learn, PyTorch, pandas) and service frameworks (FastAPI, Pydantic).
• Experience with testing, code review, CI/CD, and production ownership.
• Strong hands-on experience with ML evaluation techniques including time-aware validation, backtesting, holdout design, metric selection, calibration, and error analysis.
• Production experience with forecasting models and statistical uncertainty quantification.
• Expertise in model serving and MLOps including reproducible deployment, monitoring, retraining, experiment tracking, and model registries.
• Ability to identify and judge failure modes such as data leakage, drift, bias, privacy leakage, and silent degradation, and communicate model behavior and uncertainty to technical and non-technical stakeholders.

Preferred, nice-to-have Skills

• Experience with embeddings and retrieval systems and measuring retrieval quality.
• Knowledge of synthetic data generation, privacy attack evaluation, and differential privacy.
• Familiarity with geospatial or location data modeling.
• Experience building models callable by LLM agents using frameworks like LangGraph and LLM tracing tools.
• Experience with batch pipeline orchestration tools such as Airflow or Step Functions and self-hosted model inference.
• Domain experience in real estate, facilities, finance, or other data-heavy enterprise sectors.

About the Client and Project Context

The Senior Machine Learning Engineer will contribute to a leading cloud-based SaaS company specializing in end-to-end real estate and facilities lifecycle management. The client’s software solutions cover store lifecycle management and integrated workplace management systems, including real estate, design and construction, lease administration and accounting, facilities maintenance, occupancy management, energy and sustainability, desk booking, visitor and space management. The role supports hundreds of customers across more than 140 countries.