Senior ML Engineer

MLabs•New York, NY

1d•$300,000 - $375,000•Onsite

About The Position

Our client, a premier venture accelerator backing high-growth technology and emerging-tech startups, is seeking an experienced and self-directed Senior Machine Learning Engineer to join their in-house engineering team in New York City. Reporting directly to the Chief Technology Officer (CTO), the Senior Machine Learning Engineer will take full ownership of features throughout the product lifecycle—from requirements definition to production deployment. This onsite role requires an entrepreneurial mindset and deep technical execution to turn loosely defined problems into robust, production-ready machine learning and LLM-based systems without reliance on large engineering teams or dedicated project managers.

Requirements

Senior-Level ML Expertise: Proven ability to drive complex, ambiguous ML problems from initial experimentation through to reliable production releases.
Production Python Proficiency: Strong mastery of Python across data exploration, training pipelines, application logic, APIs, and production debugging.
Robust Modeling Judgment: Deep experience in problem formulation, label definition, feature engineering, evaluation metrics, backtesting, data leakage prevention, calibration, interpretability, and model selection.
Full-Stack ML Engineering Capability: Strong software engineering and data pipeline fundamentals to deploy, integrate, schema-manage, and monitor services autonomously.
Applied LLM Systems Experience: Hands-on experience with structured outputs, model/prompt evaluation frameworks, observability, retry strategies, cost/latency optimization, and input validation.
Product Sense & Communication: Ability to communicate technical tradeoffs clearly with non-technical stakeholders and translate model outputs into actionable business tools.
Location: Based in or willing to relocate to New York City (onsite presence is strictly required).

Nice To Haves

Track record of shipping customer-facing ML products with end-to-end ownership.
Strong portfolio of technical work (e.g., active GitHub, open-source contributions, published research, or technical writing).
Prior experience developing prediction, ranking, classification, recommendation, or anomaly-detection systems on messy, real-world data.
Background building LLM evaluation frameworks, structured extraction pipelines, or automated research agents.
Early-stage startup experience as a founder, early ML hire, or senior IC working without dedicated platform teams.
Exceptional technical or quantitative pedigree (e.g., strong research background, competition achievements, or top-tier academic background in quantitative disciplines).
Direct partnership with high-impact founders across emerging technology sectors.
High-trust, high-autonomy environment within a lean, elite engineering team composed of industry veterans.
Direct ownership over core systems shaping founder discovery and operational workflows.
Accelerated career growth and networking opportunities within a leading startup ecosystem.

Responsibilities

Own Applied ML End-to-End: Translate ambiguous, high-level business problems into datasets, experiments, models, and production systems independently.
Build Production Pipelines: Develop and maintain production Python systems for data collection, enrichment, feature extraction, scoring, model evaluation, and AI-assisted research workflows.
Model Design & Evaluation: Define labels and features, construct evaluation sets and backtests, detect data leakage, evaluate source quality, and select optimal modeling approaches (including traditional ML and LLMs).
Production Deployment & Operations: Transition models from research to production environments, managing artifacts, feature/prompt compatibility, APIs, background jobs, observability, failure handling, and release cycles.
Enhance LLM Infrastructure: Improve large language model systems, including structured data extraction, research agents, prompt and model evaluation, and safety guardrails for untrusted external inputs.
Stakeholder Collaboration: Work directly with key stakeholders to determine roadmap priorities, clearly articulate model behavior and tradeoffs, and iterate iteratively based on real-world usage.