Python Engineer Job Spec

About Bizimply Bizimply is a workforce management platform designed to streamline operations for businesses in the hospitality, retail, and leisure industries. The Bizimply platform provides a comprehensive suite of tools that enable businesses to efficiently manage their workforce, including scheduling, time and attendance tracking, task management, HR and performance reporting. By centralising these functions in one user-friendly interface, Bizimply helps businesses save time, reduce administrative overhead, and improve overall operational efficiency. About the role Bizimply is a workforce management platform used by thousands of businesses. We are building AI-powered features that help our customers make better decisions using their workforce data, and we are looking for a Python engineer to help us build and scale that capability. You will work on a standalone Python AI service that sits alongside our main product, integrating with LLM providers, querying analytical data stores, and streaming intelligent responses to end users in real time. Location Dublin, Ireland. Hybrid: 2 days per week in the office. Key Responsibilities What you will work on Building and extending a production Python AI service LLM tool calling, agentic loops, and prompt engineering Designing and running LLM evaluations to maintain quality as features evolve Writing SQL against analytical databases to ground AI responses in real data LLM cost and usage observability Working within a defined service boundary alongside a Rails backend Skills, Knowledge and Expertise Requirements 3+ years of professional Python experience, writing production services Experience building with LLM APIs, including tool use, structured output, and prompt construction Comfortable writing SQL against analytical databases Experience with async Python and streaming HTTP Good security instincts: prompt injection, PII handling, and service-to-service auth patterns Comfortable using AI coding agents as part of your daily workflow, while knowing when to rely on your own judgement rather than the output Nice to Haves Experience with ClickHouse or other columnar analytical databases Familiarity with agentic system patterns: multi-step tool calling, context management, memory GDPR considerations for LLM systems, particularly around PII sent to third-party providers Experience with Airflow or similar data pipeline tools Some exposure to Rails or Ruby Awareness of EU AI Act requirements