Senior AI Software Engineer

Key Responsibilities

• Lead the design and development of enterprise-grade GenAI solutions, owning architecture decisions, technical patterns, and platform standards.
• Act as technical authority on LLMs, agentic systems, and emerging AI technologies, guiding adoption decisions and technical direction across engineering.
• Evaluate, prototype, and document emerging tools, platforms, and techniques — recommending strategic investments aligned with Litera's roadmap.
• Own technical estimates, solution feasibility assessments, and delivery accountability for complex, cross-functional initiatives.
• Establish and maintain GenAI governance standards, including safety, security, and responsible use guidelines applicable to production deployments.
• Define and own LLM observability and evaluation standards — including tracing, quality metrics, and monitoring for production AI systems.
• Champion software engineering best practices including secure coding, design patterns, and performance optimization.
• Contribute high-quality, scalable, and secure code with a strong focus on maintainability and reliability.

Required Experience

• 6+ years of professional software development experience, with at least 2 years focused on production GenAI systems and applications.
• Deep, hands-on experience with Large Language Models — including context engineering, prompt design, structured outputs, and LLM evaluation.
• Experience with LLM observability and evaluation tooling (e.g., LangSmith, RAGAS, Arize, PromptFlow) and the ability to define and track GenAI quality metrics in production.
• Proficiency in Python, with working experience in LangGraph and/or LangChain.
• Familiarity with agent-based and multi-agent orchestration architectures.
• Experience with enterprise SaaS RESTful APIs and cloud-based backend development.
• Comfortable with relational data stores (PostgreSQL preferred; Azure SQL or SQL Server also acceptable).
• Solid background in testing practices — including unit, integration, and evaluation-layer testing for AI systems (e.g., pytest).