Senior AI Engineer - Agentic Systems

‎

About the opportunity

Join QIAGEN Digital Insights, a global leader in bioinformatics software and knowledge solutions that help scientists transform complex molecular data into actionable biological insights. Our SaaS products — including OmicSoft, our Discovery Knowledge Base, Ingenuity Pathway Analysis (IPA), and QCI — give pharma, biotech, academic, and government lab customers worldwide the proprietary datasets, biological knowledge, and analysis algorithms they need to interpret human molecular (genomic) data. Today, customers access these capabilities primarily through our web UIs.

As an AI Engineer leading our agentic tool integration effort, your job is to change that: you will help design and build the tools and services that let AI agents — both external agentic AI accessing us via MCP, and internal AI orchestration within QIAGEN — successfully use our data, knowledge, and analysis capabilities without a UI. Authentication and access authorization for external MCP access are handled by an existing gateway layer; your work begins after that boundary. Your focus is turning UI-driven analytical workflows into clean, agent-native tools backed by our existing product APIs and services, and structuring their outputs so LLMs can reason over and act on them reliably.

You'll work closely with bioinformaticians, product owners, and application engineers, based out of our Cluj-Napoca site. We're looking for a strong Python engineer with genuine curiosity about life sciences, hands-on experience with agentic AI development, and good judgment about how to design software interfaces for AI consumers.

QIAGEN Romania offers a professional work environment, valuing work-life balance.

Your tasks:

• Knowledge and Data Integration for Agentic Systems: Integrate and expose heterogeneous scientific and product data sources — including knowledge graphs, relational databases, metadata services, analysis outputs, and existing product APIs — so they can be reliably used by LLM-based agents and MCP tools. This includes designing knowledge graph structures, embeddings, and query interfaces, as well as query/response patterns for other data sources, that make relationships, identifiers, confidence, provenance, and ambiguity explicit — ensuring agents can discover, query, and reason over the data safely and predictably, whether the underlying source is a graph, a database, or an API. 

• Model and data for agentic systems: Partner with application engineers to build MCP servers and agent-orchestration tools to ensure models, datasets, and knowledge representations can be reliably surfaced to LLM-based agents. This includes structuring outputs, preserving provenance and confidence metadata, and flagging ambiguity so agents can request clarification rather than hallucinate an answer.

• Software Engineering: Write well-documented, extensible software code that is easy to maintain and adheres to generally accepted programming standards and OOP practices.

• Collaboration: Work cross-functionally with bioinformatics experts and software engineers to align AI solutions with customer requests, downstream tool requirements, and our company’s strategic goals. 

• Responsible and transparent AI: Ensure transparent reporting of AI methodologies, data provenance, and known limitations to maintain customer and public trust. Proactively address risks like model bias, data security, and intellectual property, with particular attention to how biological claims are traced and communicated when relayed through an AI agent rather than presented directly by a human analyst.

Your profile:

• 5+ years working as a Data Scientist or AI/ML Engineer, with experience turning AI/ML methods into reliable, production-ready capabilities used by real users or downstream systems.

• Strong Python development skills, including asynchronous programming (async/await) and building REST APIs/microservices with frameworks such as FastAPI.

• Hands-on experience building or integrating with at least one agentic AI framework (e.g., LangChain, LangGraph, or the MCP SDK), and direct experience building MCP servers

• Expertise in data preprocessing, feature engineering, and model evaluation techniques

• Experience integrating foundation-model APIs into applications or agentic workflows, including structured output design, retrieval/tool interfaces, and metadata such as confidence, provenance, or citations to support reliable downstream LLM reasoning.

• Ability to own quality and evaluation for agentic AI systems, including tool-call correctness, structured-output validation, retrieval quality, provenance handling, regression testing, and failure-mode analysis.

• Familiarity with source control and issue tracking systems and CI/CD workflows

• Understanding of Deep Learning algorithms

Nice to have:

• Developed full data-focused applications to analyze and process data

• Worked with biological data (e.g., gene expression profiles, pathways, gene metadata, etc.)

• Proficiency in graph-based analysis and network modeling, including experience with tools like NetworkX and Neo4j

• Research experience in bioinformatics, AI, or drug discovery