Principal Machine Learning Engineer

This is a fully remote position, open to applicants in Massachusetts.

📋 Description

• Develop the technical strategy for the data graph, entity resolution, buying intent, and agent context.

• Identify and deliver innovative machine learning challenges, such as extraction for companies lacking a public presence and semantic intent analysis at web scale.

• Establish evaluation criteria, including sampling methods, human labels, validated LLM judges, and release protocols.

• Create data agreements between the data graph, intent signals, and consuming agents, incorporating agent memory.

• Ensure systems remain cost-effective across the entire dataset.

• Set organization-wide defaults for model baselines, uncertainty reporting, and performance metrics.

• Conduct design reviews and provide mentorship to senior engineers.

• Influence hiring decisions and advocate for the team’s technical expertise to product and leadership stakeholders.

⛳️ Requirements

• Provide technical leadership as an individual contributor for a problem domain that spans multiple teams, with ownership of production machine learning systems post-launch.

• Possess a broad understanding of classical machine learning applied to large, complex tabular data, applied statistics, large-scale language processing, and LLM agents or multi-step systems, with in-depth expertise in at least one area.

• Define and promote evaluation standards beyond the immediate team for systems that do not have a singular correct answer, including LLM judges validated against human labels.

• Demonstrated ability to influence how a group constructs or measures its systems.

• Experience in making decisions regarding cost and capacity for model training and serving, including options for self-hosting, distillation, or utilizing hosted models.

• Proficient in production-level Python and strong SQL, with experience in distributed data processing.

• Regularly utilize AI coding tools with a thorough review of their outputs.

• Preferred: expertise in ranking and retrieval, embeddings, learned re-ranking, recall@k, MRR, and nDCG.

• Preferred: background in entity resolution, knowledge graphs, or web-scale information extraction.

• Preferred: experience in training and serving open-weight models using PyTorch or similar frameworks, including planning for GPU capacity.

• Preferred: familiarity with user memory or context systems for agents, or strategies for defending LLM systems against adversarial attacks and prompt injection.

🏝️ Benefits

• Comprehensive benefits package.

• Holistic programs for mind, body, and lifestyle designed to promote overall well-being.

• Additional compensation may include bonuses, commissions, and equity options.