We are looking for a Middle AI Engineer to join our team, implementing clinical solutions for pharmaceutical product development and testing. In this role, you’ll build and optimize AI agents for clinical operations using the SADK framework, with a strong emphasis on prompt engineering, agent configuration, and iterative improvements driven by user feedback.
This is you * 3+ years of software engineering experience, including 1+ years with LLM/AI technologies; * Strong prompt engineering skills; * Experience with OpenAI, Anthropic, or Azure OpenAI APIs; * Proficiency in Python or TypeScript; * Understanding of RAG (Retrieval-Augmented Generation) and vector search; * Knowledge of testing and evaluation methodologies; * Proficiency with Git version control; * English — Upper-Intermediate level or higher.
Nice-to-have skills: * Knowledge of clinical trials or healthcare domains; * Experience with agent frameworks; * Data analysis and visualization skills (e.g., Python libraries like pandas and matplotlib).
This is your role * Develop agents using SADK archetypes and YAML configurations; * Write and optimize prompts tailored for clinical use cases; * Implement agent tools and integrations via MCP; * Configure guardrails, including PII detection, grounding checks, and HITL (Human-in-the-Loop) triggers; * Test agents using evaluation frameworks and test suites; * Analyze agent performance metrics, such as accuracy, response quality, and cost; * Iterate on agents based on user feedback and production metrics; * Document agent specifications and operational runbooks.