Machine Learning & Data Platform Engineer
About The Position
We are looking for a Staff Engineer to own and evolve our financial data integration and ML classification platform. This system ingests hundreds of variations of real estate financial reports — trial balances, rent rolls, budgets, forecasts, aged receivables — from property management companies, automatically detects their format, classifies their structure using ML models, and transforms them into normalized data for downstream analytics. You will be the primary technical lead across two critical systems: a document parsing engine handling 280+ specialized ETL processors and a dynamic pipeline orchestration platform that uses ML-predicted field mappings to automate data extraction at scale. This role demands breadth — you will build ML models, maintain production data pipelines, design database schemas, and make architectural decisions independently. You will also mentor and manage one direct report.
Requirements
- 5+ years Python development with production ML systems
- Strong experience with NLP / text classification — specifically training and fine-tuning transformer models (Hugging Face, TensorFlow) for document understanding tasks
- Deep proficiency with pandas, NumPy, SQLAlchemy, and PostgreSQL
- Experience building and maintaining ETL pipelines that process messy, semi-structured data (Excel, CSV) at scale
- Familiarity with ML model serving platforms (Wallaroo, SageMaker, Vertex AI, or similar) including OAuth2-based inference APIs
- Comfort operating as a technical lead — triaging bugs, shipping features, making architectural calls independently, and mentoring junior engineers
- Demonstrated ability to work across the full stack of a data platform: from raw file ingestion through model inference to API integration
Nice To Haves
- Domain experience in real estate finance, property management, or accounting data (chart of accounts, trial balances, rent rolls)
- Experience with SFTP-based file processing workflows and Paramiko
- Familiarity with dynamic code generation / evaluation patterns for configurable data transformations
- Background in document parsing / OCR / intelligent document processing
- Experience with SSH-tunneled database connections and multi-environment deployments
- Prior experience mentoring or leading a small team
Responsibilities
- ML Model Development & Deployment — Build, train, and deploy classification models (currently served via Wallaroo) that predict financial field/table mappings from document headers. Extend existing BERT-based question-answering models used for extracting structured data from free-text property descriptions.
- Pipeline Platform — Maintain and extend the configuration-driven orchestration system that matches incoming files to processing pipelines, executes dynamic conditionals, and writes standardized output to downstream templates.
- ETL Engine — Evolve the parsing library that handles complex Excel workbooks with nested headers, multi-tab structures, merged cells, and varied accounting system formats (ARES, GIC, DCS).
- Data Quality & Reliability — Improve prediction accuracy, expand audit logging, and ensure processing integrity across clients and document types.
- Architecture & Technical Leadership — Drive technical decisions on model serving infrastructure, database schema design, and API integration patterns with RealPage's Data Management Gateway (DMG). Provide mentorship and technical guidance to your direct report.
Benefits
- Health, dental, and vision insurance.
- Retirement savings plan with company match.
- Paid time off and holidays.
- Professional development opportunities.
- Performance-based bonus based on position.
Stand Out From the Crowd
Upload your resume and get instant feedback on how well it matches this job.
What This Job Offers
Job Type
Full-time
Career Level
Senior
Education Level
No Education Listed
