Applied ML Engineer: Housing Verification & Fraud Detection

Two Dots; San Francisco (CA); On-site; USD 180,000 - 240,000; Full time

Job summary

Two Dots is hiring a Machine Learning Engineer for a low-headcount, high-impact role focused on applied ML in housing verification, underwriting, fraud detection, and document understanding.

This is not a pure research position; you will develop end-to-end models, handle difficult problems like PDF forgery detection and latent financial profile inference, and deploy reliable ML systems with BigQuery data workflows.

Qualifications

• Strong command of tensors, PyTorch, training loops, and model deployment.
• Rigorous, metrics-driven evaluation and quality management.
• Solid grounding in statistics, regularization, and GPU memory management.
• Experience with computer vision, NLP, and multimodal understanding.
• Proficient in data warehouse SQL, especially BigQuery.
• Able to balance explore-vs-exploit in applied ML work.

Responsibilities

• Document forensics and detecting fraudulent or edited PDFs.
• Cash flow underwriting: infer latent financial profiles from paystubs, bank statements, etc.
• Extract information from noisy sources with high reliability.
• Solve chatbot and agent quality problems that are hard for others.
• Develop models, evaluation systems, and quality processes from scratch.
• Drive broad ML/LLM improvements and pipeline education for the team.

Skills

Tensor operations

PyTorch

Training loops

Model deployment

Metrics-driven evaluation

Statistics

GPU memory management

Computer vision

NLP

Multimodal understanding

BigQuery

Explore-vs-exploit

Job description

Two Dots is hiring a Machine Learning Engineer for a low-headcount, high-impact role focused on applied ML in housing verification, underwriting, fraud detection, and document understanding.

This is not a pure research position; you will develop end-to-end models, handle difficult problems like PDF forgery detection and latent financial profile inference, and deploy reliable ML systems with BigQuery data workflows.