AI Engineer (LLM)

About dexter health

dexter health develops AI-driven software for care teams, with the goal of cutting down administrative work in healthcare so caregivers can focus more of their time on patients.

Role overview

They are hiring a hands-on AI Engineer who can move from an unclear product idea to a working, production-grade AI feature. This position is centered on building and shipping AI product functionality rather than doing research or only writing prompts. The work is mostly about delivering AI features, with some related systems and infrastructure tasks where needed.

You will partner closely with product and engineering, but the expectation is that you will take real ownership, make sound technical calls in ambiguous situations, and use AI coding tools such as Claude Code, Codex, Cursor, Copilot, or similar tools as part of your daily workflow.

What you'll do

• Create new AI-powered product capabilities from concept through production release.
• Refine current AI workflows to improve accuracy, reliability, speed, and overall user value.
• Design LLM-based pipelines with structured responses, validation rules, and fallback logic.
• Build test harnesses, evaluation loops, and quality checks for AI-generated results.
• Connect AI features to the existing product and backend architecture.
• Work with both commercial and open-source language models without depending on a single vendor.
• Help run self-hosted model setups when they are the better choice for quality, performance, cost, or control.
• Investigate and fix failures across prompts, inputs, outputs, backend logic, data, and user journeys.
• Use AI development tools as a regular part of day-to-day engineering work.
• Move quickly while still maintaining a high bar for production quality.

Requirements

The company values speed, judgment, ownership, and product impact more than formal seniority. They are looking for someone who works with founder-like urgency, can make progress without perfect instructions, and cares more about outcomes than job titles.

• Strong software engineering ability, especially in Python or another backend language.
• Proven experience building AI product features or LLM-based workflows.
• Ability to convert vague product ideas into working software quickly.
• Solid understanding of how to design, test, verify, and improve AI outputs.
• Experience bringing AI workflows into production systems.
• Strong debugging skills across application logic, data, model behavior, and user-facing issues.
• Experience with evaluation, testing, validation, or quality assurance for AI-generated outputs.
• Hands-on use of AI development tools such as Claude Code, Codex, Cursor, Copilot, or similar tools on a daily basis.
• High learning speed, strong initiative, ownership, and bias toward action.
• Comfort working in ambiguity, making fast decisions, and operating in a high-trust startup setting.
• Clear written and spoken English.

Preferred experience

• Exposure to frontend or full-stack product development.
• Experience with self-hosted LLMs, model serving, inference tuning, or similar infrastructure.
• Work with open-source LLMs and their deployment processes.
• Background in improving AI system latency, cost, quality, or reliability.
• Experience in healthcare software, healthcare data, or regulated environments.
• Experience with structured data extraction, documentation automation, voice-based workflows, or workflow automation.
• German language ability.
• Familiarity with nursing home or elderly care workflows.

Not a fit if

• You are mainly seeking an ML research position.
• You are a prompt engineer without strong coding ability.
• You need highly detailed instructions before you can make progress.
• You prefer slow, rigid, corporate-style environments.
• You are looking primarily for a management role rather than hands-on building.
• You only want to use AI tools occasionally instead of every day.
• You prioritize perfect process over fast, dependable execution.

Benefits

• Remote work.
• Compensation that reflects skills, experience, and location.
• Ownership of key backend systems.
• Opportunity to build software used in real healthcare operations.
• Modern AI-native development setup.
• Room for career growth as the company expands.

Why this role exists

Care teams spend too much time on documentation, coordination, and other administrative tasks. The company is building technology that helps reduce that burden.

AI is central to the product, but effective AI features require more than polished demos. They need strong engineering, rapid iteration, evaluation, production thinking, and deep ownership of user outcomes.

If you want to build AI product features that ship, improve real workflows, use AI tools every day, and take ownership in a fast-moving team, this role is designed for you.