Principal AI/ML Engineer

1 мес. назад

240k–320k USD / yearUSALeadRemote

pythonmachine learningdeep learningdata sciencetensorflowpytorch

Principal AI/ML Engineer role building an audited methodology platform and marketplace for rules-based investment strategies, combining human and AI agents.

О компании

We're a team of ~25 senior-to-principal engineers, designers and AI/ML researchers. More than half the team are former founders. We move fast, challenge assumptions, and believe AI should fundamentally change how financial products are built. We have no management hierarchy and work remotely with 4 one-week hackathons per year. “Full stack” to us means you own a slice of the product, including its roadmap and go-to-market.

Обязанности

Creating an investable index from a conversation with AI isn't a traditional machine learning problem. Users aren't simply prompting a chatbot—they're collaborating with an intelligent system that must reason over financial information, construct investment methodologies, evaluate evidence, and continuously improve as new information emerges.
We're looking for a Principal AI/ML Engineer who thrives in ambiguity, enjoys solving complex research and engineering problems, and wants to define what AI-native investing feels like.
You'll work directly with the founders, product engineers, and designers to shape both the intelligence behind the product and the company itself. This isn't a role where you'll be handed research problems. You'll help identify opportunities, influence our AI strategy, and set the standard for machine learning across .
Design and build the AI systems that power methodology generation, financial reasoning, retrieval, ranking, and evaluation.
Develop agentic workflows that combine language models, structured financial data, and external tools into reliable production systems.
Improve the quality, consistency, and trustworthiness of our AI through evaluation frameworks, benchmarking, and continuous experimentation.
Partner closely with product engineers and designers to turn cutting-edge AI capabilities into intuitive product experiences.
Stay at the frontier of modern AI, rapidly evaluating new models, techniques, and tooling as the ecosystem evolves.
Make thoughtful tradeoffs between research ambition, production reliability, latency, and cost.
Raise the AI bar across the company by contributing to architecture, experimentation, and technical leadership.
Continuously improve how we build by leveraging AI throughout research, experimentation, coding, and evaluation.

Требования

You've spent years building production AI systems, not just training models, and understand the difference.
You naturally think about product outcomes alongside model performance.
You have strong machine learning and software engineering fundamentals and are comfortable taking ideas from research through deployment.
You're able to decompose difficult research problems into practical, scalable systems.
You know when to pursue state-of-the-art techniques and when simpler solutions deliver better outcomes.
You communicate technical ideas clearly and enjoy working closely with engineers, designers, and founders.
You're comfortable taking ownership without waiting for direction and enjoy exploring entirely new problem spaces.
You believe AI should fundamentally change how financial products are built and actively experiment with the latest models and tooling.

Условия

Competitive base salary and meaningful equity package
Direct partnership with the founders and a team where more than half are former founders.
Comprehensive health and dental benefits.
A remote-first, flexible work environment.
The opportunity to help define a company creating a new category in capital markets.

Другое

Agora Intelligence is building the audited methodology layer agents call before touching portfolios. Our product, , is GitHub for rules-based investment methodologies: an open platform and marketplace where investment strategies are created, tested, audited, and distributed — by humans and AI agents alike. As AI agents begin managing real portfolios, they need a trusted source of verified investment logic before they can act. is that source.
We operate across two connected surfaces: InvestOS , an open standard that lets any brokerage or platform plug verified methodology into its own AI experience; and , the reference implementation of InvestOS. is an agentic harness where investors, portfolio managers, and creators can author, remix, and publish strategies. We're a small, senior team (from BlackRock, Bloomberg, BGI, and Quant Funds) backed by Portage Ventures, Lerer Hippeau, Cumberland Investments (and more), working with major brokerages and financial data providers.
InvestOS — an open standard for rules-based methodologies and its reference implementation.
We believe any idea — a theme, a trend, a thesis — can be turned into an investable view. On , anyone can express a view as a rules-based basket of assets in seconds, backed by evidence and scored against what actually happens. Whether you're modeling a thematic strategy, tracking an emerging technology, or capturing the public-market footprint of a private company, what used to take an index committee months now takes an API call — 100× faster, at a fraction of the cost. We also transform views found on the open web, into rules-based strategies. This lets us measure the performance of any opinion or prediction about the future. The unit is a : a creator-authored view of the world.
The bigger bet: as AI agents take on more of investing, they'll need an open, trusted standard for how investment views are expressed, verified, and combined. When anyone can generate a strategy in seconds, the strategies aren't the asset. The system that versions, tests, and grades them is. We're building that standard — and the platform where those views are made. We’re also building the learning loops for when views resolve against reality, as measured by markets.
Experience building production systems with large language models, agents, or retrieval-augmented generation.
Experience designing evaluation frameworks for AI quality, reliability, and reasoning.
Strong software engineering skills alongside deep machine learning expertise.
Experience working with financial, scientific, or other information-dense domains.
Experience at an early-stage startup where you've owned significant technical or research decisions.
You're energized by difficult AI problems and continuously explore new models, research, and infrastructure.
A link to your GitHub, publications, portfolio, or examples of your work.
A short note about an AI system you've built that you're most proud of, including the problem you solved and your role in the outcome.
A few sentences on why you're interested in .