# Lead AI Engineer, LLM Systems

**Company:** [hermeneutic research](https://jobs.workable.com/companies/qFgeH84KbBLAfWded89WjF.md)
**Location:** Remote
**Workplace:** remote
**Employment type:** Full-time
**Department:** Technology

[Apply for this job](https://jobs.workable.com/view/5bb9ddf8-aca9-4a5c-bef7-eefda9c7860d)

## Description

### **Company Overview**

hermeneutic research is a best-in-class proprietary trading firm. It deploys research-driven discretionary and systematic strategies as well as makes strategic long-term investments. The partners' decade-long history of success in trading and business building and a firm-wide cultural emphasis on alpha generation, open debate, relentless iteration, and teamwork are key to the firm's continued expansion in a challenging market environment that has hamstrung competitors. A hard-wired emphasis on risk management and opportunistic market participation ensures that hermeneutic research will continue its growth trajectory in the coming decades.

### **Job Overview**

You will own a multi-agent research system as production infrastructure: make it faster, cheaper, measurably better and efficient.

hermeneutic Research Labs is the firm’s applied research and engineering unit. Its first product is a research intelligence platform for both traditional and digital-asset markets, built with our researchers and traders and used by them every day. The platform turns market, on-chain, macro and news data into event-driven research that trading decisions rely on. The platform is held to the standard we apply to our own trading: rigorous, falsifiable and measured on outcomes.

The Labs is making a multi-year investment to turn the platform into production infrastructure that a growing set of internal users can depend on. As part of that investment, it is hiring a Lead AI Engineer.

The LLM layer is yours for years. A research query runs through multiple agent stages across several model providers and our own self-hosted models. You start with your hands on the system, writing code and running experiments. As the function grows, you will hire and lead its engineers. The role is built to grow into Head of AI Engineering.

### Key Responsibilities

-   Keep the LLM layer reliable: latency budgets, concurrency and queueing, declared degradation modes, structured-output validation and provider failover.
-   Build evaluation tied to research outcomes: offline sets from real queries with outcome labels, online experiments, regression gates on every change, and model-generated confidence calibrated with quantitative research before it reaches a decision.
-   Control cost and routing: per-workload cost attribution, difficulty-based routing between small and frontier models, caching and batching, and the economics of self-hosted serving and GPU capacity.
-   Run model strategy and lifecycle: pinned model versions, replayable test batteries before any change, one-step rollback, provider diversification, and fine-tuning or adapters where they pay.
-   Shape the agent architecture with evidence: ablate and retire components that do not earn their keep, and improve tool use, memory and retrieval only where they move research outcomes.
-   Ground and trace outputs: citation, hallucination controls, audit trails, and the boundary between what a model may assert and what it may only suggest.
-   Build the team: review and mentor engineers now, and hire and lead an AI engineering group as the function grows.

## Requirements

### **Must-Haves**

-   7+ years of software or ML engineering, including 3+ years shipping LLM-backed systems to real users in production, where you owned latency, cost and quality, and 2+ years leading other engineers while still shipping.
-   A track record of building a function from nothing.
-   Command of inference economics: tokens, caching, routing, structured outputs, tool use and quantisation, with cost-per-request figures for the systems you ran.
-   Measurement discipline: you instrument before changing anything, evaluate against outcomes rather than model judges, have built an eval that caught what a vibe check missed, and have retired a component with data.
-   Strong Python, including async concurrency.
-   Experience with self-hosted serving (vLLM or similar), adapters and GPU capacity planning, or an end-to-end provider migration.
-   Depth in vector and graph retrieval, built and measured, with an informed view on when retrieval is the wrong fix.

Technologies named here are what we run today, not a filter. We hire for depth and judgement.

**Desired Attributes**

-   Directness: you tell the author when their component does not earn its keep.
-   Sound judgement working alongside AI coding agents on a largely AI-written codebase, keeping ownership of what ships.
-   Initiative: you take on the hard, unowned problem first, and bring the profiling data and a first fix.
-   Bonus: fine-tuning and preference optimisation in production, and an informed view on when multi-agent frameworks are over-built.
-   Bonus: finance or trading exposure, and experience hiring engineers.

### **Interview Process**

1.  **CV Screening** – We will review your application based on the qualifications and experience outlined above.
2.  **Hiring Manager Screen** – A conversation about the LLM systems you have run in production.
3.  **Systems Deep-Dive** – A whiteboard session on LLM systems design.
4.  **Hands-On Round** – A take-home on a synthetic trace set with a review, or a live debugging session.
5.  **Reliability and Measurement Scenarios** – A scenario discussion with peers from research or risk.
6.  **Collaboration and AI-Assisted Work** – A conversation on how you work with teammates and AI coding agents.
7.  **Co-Founder Interview** – A concluding discussion to assess your capabilities, strategic thinking, and cultural fit.

### **Company Values**

Throughout the process, you'll be assessed for cultural fit through our company values:

-   **Drive** – We believe the best team members are deeply passionate about what they do. That passion fuels their growth, drives them to seek out the best teams, and leads them to hold high expectations of themselves in pursuit of excellence.
-   **Ownership** – We aim to extend ownership as broadly as possible across the firm. In return, we value people who take initiative, step up when needed, and treat the company’s goals as their own.
-   **Judgment** – We value those who see the big picture and focus on what truly drives impact and results. They use time wisely, adapt across domains, and always prioritize outcomes over comfort zones.
-   **Openness** – We foster a culture of open communication, mutual challenge, and shared growth. We believe constructive debate and proactive knowledge sharing lead us closer to the truth—and better decisions.
-   **Competence** – We work with people who bring exceptional intellectual strength. What sets them apart isn’t just what they know—it’s how they use what they know to navigate change, adapt quickly, and contribute meaningfully in fast-changing environments.
-   **Resilience** – We perform under pressure. Markets move fast, and so do we—but we stay grounded, focused, and calm. We embrace uncertainty, learn from setbacks, and adapt quickly without losing sight of long-term goals.
