Software Engineer (Production Informatics)

Predicine, Inc•Hayward, CA

64d

About The Position

Hands-on software engineer responsible for the design, development, and reliability of Predicine's production pipeline operations and data-related applications. Owns architecture and delivery of full-stack features across front-end, service, and API layers; integrates applications with enterprise, product, and scientific data sources; and leads the adoption of AI and agentic AI capabilities — including LLM-based tooling, tool orchestration, retrieval, evaluation, and guardrails — to improve workflow automation, developer productivity, and user experience. Establishes reusable engineering standards and patterns, including for AI-assisted development, mentors other engineers, and partners directly with product, science, data, and IT stakeholders to convert ambiguous business needs into secure, well-tested, production-grade software.

Requirements

Bachelor's degree in Computer Science, Software Engineering, Data Science, or related field; equivalent practical experience considered.
7+ years of professional software/application development (full-stack, backend, or similar), including delivery and support of production, data-rich applications; demonstrated technical ownership and mentoring.
Strong Python and REST API development (e.g., Django/DRF or equivalent).
Working knowledge of TypeScript/modern web stacks (React).
Data integration patterns (permissions, lineage, auditability).
Working familiarity with AI coding/productivity tools (e.g., Cursor, ChatGPT, Claude).
Git, CI/CD, containerization (Docker), SQL, cloud deployment, automated testing, secure coding practices.
Willingness to complete Predicine-provided training in agentic AI/LLM development.
Experience integrating heterogeneous data sources and structured/unstructured data with clear grounding, citations, and auditability.
Strong bias for automation, reproducibility, and secure, well-tested delivery.
Comfortable operating with general supervision, exercising independent judgment, and mentoring less-senior engineers.
Applicants must be legally authorized to work in the USA without sponsorship.

Nice To Haves

Master's degree (or equivalent leadership/technical experience).
Applied AI/machine learning coursework or focus preferred.
10+ years of experience.
Hands-on experience building or integrating LLM-based or agentic AI features in a production application, ideally in life sciences, healthcare, diagnostics, or another regulated data environment.
Cloud certifications (AWS/Azure).
Security/compliance training (HIPAA/GxP aware).
Demonstrated, hands-on experience designing or building agentic AI systems (tool calling, orchestration, memory, retrieval, evaluation, guardrails).
Experience with reusable AI platforms, workflow engines, vector/search systems, and event-driven services.

Responsibilities

Design, develop, and maintain software solutions that support laboratory operations and production workflows.
Enhance laboratory information management systems (LIMS), backend services, databases, APIs, and user-facing applications to improve data accuracy, traceability, efficiency, and scalability.
Translate operational requirements into reliable, secure, and compliant solutions.
Troubleshoot production issues, improve system performance, and automate manual processes.
Ensure informatics systems meet business and regulatory requirements.
Own the design, development, and technical direction of key production pipeline operations and data-related applications across front-end, service, and API layers.
Drive productionization (resilient APIs, error handling, permissions, versioning, and deployment).
Design, build, and productionize LLM-based and agentic AI features (tool calling, orchestration, retrieval, memory, evaluation, guardrails).
Apply AI/agentic AI tooling where it meaningfully improves workflow automation, developer productivity, or user experience.
Connect applications to enterprise, product, and scientific data sources through APIs and well-defined data contracts.
Implement data access, indexing, and structured-output patterns.
Partner with data owners on readiness, lineage, permissions, and reuse.
Create automated tests and quality checks (including for AI-enabled functionality).
Implement tracing, observability, feedback capture, and regression testing.
Enforce access control, audit logging, privacy controls, and secure-development practices.
Establish and evangelize reusable engineering patterns, including for AI-assisted development.
Lead code review and quality standards.
Mentor engineers on best practices.
Partner with product, science, data, and IT stakeholders to prioritize work, define delivery roadmaps, and convert ambiguous requirements into reliable software.
Triage production incidents across the application portfolio.
Drive root-cause analysis and corrective action.