Clinical AI Data Engineer

Type

Full-time

Ascend Technologies Group is a U.S.

-based provider of managed IT and cloud services, specializing in telecom and data solutions. We serve clients across the U.S. with tailored strategies for enterprise accounts, focusing on cost optimization, proactive monitoring, and efficient resource use.

We emphasize precise, measurable outcomes delivered on time and within budget, supported by 24/7 technical assistance, network operations, and advanced data management. Our mission is to empower businesses through innovative technology, prioritizing security and efficiency to ensure data safety and accessibility. Our core values include excellence in solutions, innovation to optimize ecosystems, professionalism through respect and integrity, intentional action with clarity, impactful growth and accountability, and strategic expertise for precise results.

We re seeking a Clinical AI Data Engineer to build production-grade LLM systems that extract, structure, and validate cancer-related data from electronic health records. You ll work at the intersection of oncology, applied AI, and clinical informatics—developing intelligent agents that transform unstructured clinical documentation into research-ready datasets. This role emphasizes deep understanding of oncology data structures and advanced LLM techniques over traditional data engineering.

• Build Production LLM Systems for Oncology Data Extraction Design and deploy AI systems that reliably extract oncology-specific information from clinical notes and reports, including staging classifications, biomarker results, treatment regimens, and patient and outcomes.
• Develop Robust AI Agents for Medical Reasoning Create AI agents that handle complex clinical tasks: multi-document synthesis across patient charts, precise entity and relationship extraction for cancer phenotypes, and long-context understanding of treatment histories spanning years. Navigate the nuances of oncology terminology and clinical reasoning patterns.
• Ensure Clinical Accuracy and Reliability Develop strategies to minimize hallucinations, improve factual consistency, and gracefully handle ambiguous or incomplete clinical documentation. Build evaluation frameworks that measure precision, recall, and clinical validity against gold-standard oncology annotations.
• Master Oncology EHR Data Structures Work deeply with cancer-specific EHR data including pathology reports, radiology imaging summaries, genomic test results, and treatment documentation. Understand relationships between diagnosis codes, medication orders, lab values, and clinical narratives.
• Drive Technical Excellence Through LLM Experimentation Establish benchmarking standards and evaluation metrics for clinical NLP models. Experiment with advanced prompting techniques, retrieval-augmented generation, fine-tuning approaches, and multi-agent architectures. Conduct hands-on analysis to identify edge cases, model drift, and opportunities for improvement.
• Bridge Clinical and Technical Domains Collaborate with oncologists, clinical data abstraction leads, and product managers to translate complex clinical requirements into technical solutions, iterating based on real-world feedback.

• 5+ years writing production Python code with proven experience shipping AI/ML systems to production
• 3+ years hands-on experience with LLMs —including advanced prompt engineering, function calling, agent frameworks, retrieval strategies, fine-tuning, systematic failure mode analysis, and developing intuition for achieving reliable results in production environments
• Deep oncology EHR expertise
  :
  Strong understanding of cancer EHR data structures, oncology terminologies (ICD-O, SNOMED, AJCC staging, RECIST criteria), clinical documentation workflows, and how oncology data flows through health systems
• Advanced NLP and semantic understanding
  :
  Deep expertise in information extraction, entity recognition, relationship mapping, and clinical NLP challenges specific to cancer care documentation
• Production AI agent experience
  :
  You ve built agents that work reliably in real-world clinical environments, not just demos—with practical experience handling multi-step reasoning, tool use, and error recovery
• Clinical reasoning skills
  :
  Ability to understand oncology treatment pathways, interpret clinical notes, and recognize clinically meaningful patterns in unstructured documentation
• Scientific approach
  :
  You formulate hypotheses, design rigorous experiments, and iterate based on empirical evidence
• Clear communicator
  :
  Can explain complex technical tradeoffs to both clinical stakeholders and non-technical audiences
• Ability to thrive in a fast-paced, collaborative, and remote-first environment

• Experience with Snowflake or similar data warehouse platforms
• Direct experience working with oncology EHR data
• Familiarity with NGS data interpretation, and precision oncology concepts
• Experience with knowledge graphs and…