
Lead Software Engineer
Advita Ortho8 days ago
Gainesville, FL, USAStaff+
Responsibilities
- Collaborate with business stakeholders to define requirements, identify AI opportunities, and co-develop solutions.
- Assess whether large language models, conventional software, or data solutions are appropriate for each problem.
- Design and build RAG solutions involving embeddings, vector search, orchestration, and model or service selection.
- Build release-gating evaluation and production monitoring for solution quality and cost.
- Design auditable solutions in partnership with quality and regulatory stakeholders.
- Build integrations and data pipelines connected to enterprise systems of record.
- Own Azure cloud architecture, cost management, security, and access control.
- Set technical direction, write production code, support live solutions, and partner with enterprise applications teams.
Requirements
- Bachelor’s degree in Computer Science, Engineering, a quantitative field, or equivalent practical experience.
- At least 5 years of experience building and shipping production software.
- Experience delivering production applications using large language models, including RAG, prompt and context design, and evaluation.
- Experience integrating enterprise systems of record and working with their data.
- Experience with Python, Node.js, TypeScript, vector databases, and embedding models.
- Hands-on Azure experience, including Azure AI Foundry, data services, compute, identity, and networking.
- Ability to make sound judgments about when to use large language models and when to use other approaches.
- Ability to manage model, token, and infrastructure costs throughout solution design and operation.
- Strong written and verbal communication, stakeholder collaboration, multitasking, and project management skills.
- Preferred experience building and deploying machine learning models on proprietary data.
- Preferred experience in a regulated industry such as medical device or life sciences.
- Preferred experience with document-heavy and unstructured data problems, including extraction, classification, and coding.