Qualifications

The ideal candidate will possess a strong background in AI engineering, software development, and systems integration. A bachelor's or master's degree in Computer Science, Artificial Intelligence, Data Science, or a related field is required. Proven experience in building AI workflows, retrieval systems, and agent-based architectures is essential. Familiarity with programming languages such as Python and TypeScript, along with experience using Docker, Git, and CI/CD pipelines, is necessary. Knowledge of enterprise security standards, data handling, and deployment patterns is also important. Candidates should demonstrate excellent problem-solving skills, the ability to work collaboratively in multidisciplinary teams, and a proactive approach to innovation. Experience working within legal or regulated environments, understanding of AI evaluation and observability, and a strong focus on code quality and testing will be highly valued.

Responsibilities

• Develop comprehensive AI-enabled product features, including user interfaces, backend systems, AI orchestration, retrieval mechanisms, and integrations with enterprise systems.
• Design and implement agentic AI workflows utilizing structured outputs, tool calling, document processing, human-review steps, and appropriate controls to ensure safety and compliance.
• Create reusable agent harnesses that incorporate runtime patterns around tools, state management, context, approvals, and testing fixtures to ensure consistency and safety of AI agents.
• Build grounded retrieval and context-aware patterns using approved data sources such as search indexes, document repositories, and permission-aware sources, providing evidence and citations where applicable.
• Design and operate evaluation harnesses for AI workflows, including curated test sets, scenario testing, automated and human grading, regression testing, and user feedback loops.
• Manage prompts, model configurations, tool schemas, and routing strategies, including fallback mechanisms and optimization for cost and latency.
• Implement AI observability through comprehensive traces, logs, metrics, and evaluation results to monitor quality, reliability, failure modes, latency, and costs, facilitating continuous improvement.
• Develop clear APIs and integrate AI solutions with approved data, document, search, and business systems to ensure seamless operation.
• Create user interfaces that effectively communicate AI outputs, evidence, and controls, enabling users to understand, review, and provide feedback on AI-generated content.
• Write maintainable, efficient code in Python and TypeScript, leveraging Docker, Git, automated testing, and CI/CD pipelines to streamline development, testing, and deployment processes.
• Work closely with security, authentication, data handling, deployment, and release teams to adhere to firm-wide standards and best practices, ensuring the safety and compliance of AI solutions.

Initial Focus

• Develop end-to-end agentic AI features and reusable harnesses for Radar and other designated R&D products.
• Establish a practical evaluation and testing framework that enables the team to compare prompts, models, tools, and retrieval approaches both pre- and post-release.
• Create grounded AI workflows that integrate firm data, documents, market or regulatory information, and user context into reliable, attributable outputs.
• Design interfaces that clearly present AI outputs, confidence levels, supporting evidence, and next steps to legal professionals and business users.

What Success Looks Like

• Our R&D products utilize AI workflows grounded in trusted context, with operational observability and user comprehension at the forefront.
• AI quality is quantitatively measured, with evaluation results, traces, and user feedback driving continuous enhancements in usefulness and reliability.
• AI solutions are seamlessly integrated into existing systems, providing tangible improvements in efficiency, accuracy, and user experience.
• The team maintains high standards of security, performance, and maintainability, ensuring AI products are robust and scalable.