Who You Are:

You're a product-minded software engineer who turns customer problems into reliable AI-enabled capabilities. You've built and shipped production LLM features, and you know the difference between a convincing demo and a dependable product. You work across the stack, prototype quickly, validate both the value and the correctness of what you build, and turn the experiments that work into production software. You know when AI is the right approach and when conventional software is better. You're comfortable with ambiguous requirements and work with engineering, data science and customer-facing teams to find where AI creates real value for transportation agencies.

Expected Duties:

• Work with product, design, Customer Success and data science to understand agency problems, and help find and validate where AI adds value and where it doesn't.
• Design, build and ship AI-enabled capabilities within our products, end to end, from data access to the application.
• Fit model-based features into our existing software and data architecture without building unnecessary AI infrastructure, and record the key design decisions.
• Design and maintain evaluations for AI features, including correctness, grounding, authorization, appropriate abstention and regression detection. Establish practical release criteria and use production feedback to improve quality.
• Make AI-generated results traceable to their underlying data, queries and assumptions.
• Enforce customer isolation and each user's own permissions, including when tools retrieve or query data, and guard against prompt injection.
• Own features in production: latency, cost, reliability, monitoring, failure modes and incident response.
• Choose the right technique for each problem (prompting, tool use, retrieval, text-to-SQL, fine-tuning or no AI at all), weighing accuracy, cost, speed and time to market.

Required Skills & Experience:

• Bachelor's degree in Computer Science, Data Science, or a related field
• 5+ years building production software, including shipped LLM features used by customers
• Demonstrated ability to take an ambiguous product problem through technical discovery, implementation, evaluation and production delivery
• Has designed and run evaluations for LLM output: test sets, correctness and grounding measures, regression gates, and catching wrong answers before customers do
• Full-stack product skills: Python or TypeScript back ends, SQL and relational databases (PostgreSQL), APIs, and enough front end to ship a feature end to end
• Production practices: code review, CI/CD, testing, monitoring and cost tracking
• Uses AI coding tools and agents effectively in daily engineering work, including implementation, testing and investigation, while independently validating their output
• Clear written English and solid overlap with US Eastern hours

Preferred Qualifications:

• Snowflake, AWS (including Bedrock), or LLMs over analytical or structured data
• Geospatial, transportation, public-sector or other complex, high-stakes data
• Security and compliance work around customer data (multi-tenant isolation, authorization, audit, SOC 2)
• Has led a feature end to end across teams