ai engineer in healthcare

Doctolib; Berlin; Vor Ort; EUR 90.000 - 130.000; Vollzeit

Benefits dieser Stelle

Deutschlandticket

28 дней отпуска

Гибкая удалёнка

Страхование здоровья Allianz

Zusammenfassung

Doctolib разрабатывает облачную платформу и инструменты для улучшения взаимодействия между пациентами и медицинскими специалистами, поддерживая веб и мобильные интерфейсы. Команда фокусируется на проектировании и эксплуатации продакшн-агентных LLM-систем, обеспечении безопасности, наблюдаемости и улучшении рабочих процессов в здравоохранении.

Ожидается опыт работы со сложными системами AI, знание Python и практик CI/CD, а также умение работать на английском языке в ежедневной командной работе.

Qualifikationen

• 6+ лет опыта в роли инженера ПО с значительным опытом Python.
• Опыт полного цикла AI- продукта на масштабе, от оценки до продакшн-итераций.
• Понимание и внедрение лучших практик разработки ПО, включая тестирование, CI/CD и DevOps.

Aufgaben

• Разрабатывать, внедрять и эксплуатировать продакшн-ориентированные агентно-ориентированные LLM-системы.
• Проводить оффлайн-оценку гипотез и A/B- тестирование на реальном трафике.
• Сотрудничать с ML-командами для проектирования eval-пайплайнов, охраны безопасности и метрик.

Kenntnisse

Python

AI-systems

CI/CD

Testing

DevOps

English

Mentoring

Pair programming

Jobbeschreibung

Описание:

Doctolib develops healthcare products and features that improve the daily lives of care teams and patients. Its cloud-native platform supports web and mobile interfaces, multiple languages, and country- and specialty-specific healthcare requirements.

Задачи:

Design, ship, and operate production-grade agentic LLM systems that improve patient-practitioner interaction and streamline healthcare workflows; Validate hypotheses with offline evaluation datasets and A/B tests on real traffic; Partner with machine learning teams to design and improve evaluation pipelines, security and safety guardrails, and metrics; Develop and maintain core agentic capabilities, including memory, reasoning orchestration, and tool adapters, with strong observability, fallback mechanisms, and fail-safe behaviors; Own performance, observability, and security, including monitoring dashboards, SLO/SLI, and security posture; Ensure privacy and security by design; Contribute to cross-functional initiatives shaping Doctolib's future AI platform and best practices for building agentic systems.

Требования:

6+ Years of experience as a software engineer with significant Python experience; Experience with the full lifecycle of an AI product at scale, from initial evaluation and launch to production iterations based on results; Understanding and promotion of software development lifecycle best practices, including automated testing, CI/CD, and DevOps practices; Product orientation and ability to guide teammates in balancing business needs with technical constraints; Strong focus on code quality and a rigorous testing culture; Willingness to share knowledge and support team growth through pair programming and mentoring; Comfortable speaking and writing in English in day-to-day team work; Nice to have: experience with agentic systems in production, experience in the medical domain.

Условия:

A Deutschlandticket fully paid for by Doctolib; 28 Vacation days plus one additional day for each full calendar year of employment, up to 30 days; Work from abroad for up to 10 days per year through the flexibility days policy; Company health insurance with supplementary benefits through Allianz; Company pension scheme through Allianz with a 40% employer subsidy, or 15% during the probationary period; The Doctolib Parent Care program, including one additional month of parental leave; Free mental health and coaching services through Moka.care; Subsidized sports membership through Urban Sports Club; Healthy snacks, a regular breakfast buffet, and a subsidized meal benefit; Adapted remote policy, extra medical leave days, and psychological support for caregivers and workers with disabilities; The interview process includes a Recruiter Interview, Feature Building Interview, AI System Design Interview, Behavioral Interview, reference check, and offer; Start date: as soon as possible.