Founding Machine Learning Engineer

About Poesis
Poesis is the AI-native investment manager pioneering a new foundation model for investing in U.S. equities. We're building modular AI systems to predict market movements and outperform legacy managers. This is frontier research with immediate real-world validation. Your work will directly shape investment decisions and portfolio performance.

Location & Workstyle
San Francisco Bay Area (near Stanford). Hybrid: several days on-site per week.

Relocation available.

Poesis is building an AI-driven hedge fund focused on reshaping how trading decisions are made. We’re hiring our Founding ML Engineer, the first full-time machine learning hire who will turn research and data into production models.

You’ll build the first ML pipelines end-to-end — from ingesting and cleaning data, to model training, validation, and signal generation. This is a deeply hands-on, execution-oriented role for someone who can write code, design experiments, and deliver validated results quickly.

You’ll work directly with the CEO and Chief Scientist, owning both implementation and iteration. Over time, you’ll help scale the system into a full production platform and define best practices for future hires.

• Architect, build, and maintain the core ML infrastructure for Poesis’ investment platform.

• Develop reproducible pipelines for data ingestion, feature generation, and model training.

• Implement backtesting and evaluation frameworks with clear performance metrics.

• Deliver regular, documented reports on model accuracy, feature importance, and portfolio-level impact.

• Collaborate closely with the Chief Scientist to refine model hypotheses and production readiness.

• Maintain code quality: version control, testing, reproducibility, and documentation.

• Build robust backtesting frameworks and model validation tools with walk-forward evaluation and risk controls.

• Integrate with professional financial data providers (Bloomberg, FactSet, Refinitiv, CapIQ).

• Establish foundational MLOps practices: model versioning, CI/CD, monitoring, and documentation.

• Define and iterate on “demo-able” workflows that connect model outputs to investment decision-makers.

• 5–10+ years of experience as an ML Engineer, Quant Engineer, or similar role.

• Proven track record deploying production ML systems (ideally in finance or other high-stakes domains).

• Deep expertise in Python and ML frameworks (PyTorch, TensorFlow, scikit-learn, XGBoost).

• Experience designing large-scale, reliable data or MLOps systems.

• Strong software engineering fundamentals: testing, versioning, CI/CD, and code review discipline.

• Experience with financial data APIs and real-time data handling.

• Comfortable working directly with executives and acting as both IC and product owner.

• Willingness to work in-person in the Bay Area; relocation support available.

• Prior experience at a hedge fund, quant research lab, or fintech startup.

• Familiarity with quantitative finance, portfolio optimization, or risk management.

• Exposure to time-series modeling, forecasting, or reinforcement learning.

• Experience with LLM/RAG workflows for parsing financial documents (filings, transcripts).

• Comfort with multi-language engineering environments (C++, Rust, Go, etc.).

• You’re a founder-type engineer — equally comfortable writing code, setting strategy, and defining requirements.

• You thrive in high-autonomy, low-process environments and like being close to decision-makers.

• You think like both a researcher and a builder, able to turn models into production systems quickly.

• You’re pragmatic: you deliver something useful fast, then refine it as data and users evolve.

• You want to build the technical backbone of a next-generation hedge fund from day one.

Benefits: High quality dental, vision, and health care

Current legal authorization to work in the US required; visa sponsorship considered later for full-time employees.