AI Engineer (CCC Auto / Agentic)
$170k - $195k • Chicago, IL or Remote • Full-time
Posted 16d ago
About the job
As an AI Engineer at CCC Intelligent Solutions, you will be responsible for building and deploying LLM-powered features and agentic workflows for customers. These features aim to automate routine claim processes and guide customers through the remaining steps with accuracy and reliability. Agents will operate within deterministic workflows, with privacy and security as fundamental design principles. You will leverage agentic coding tools for rapid development, focusing on system design, security, and complex debugging. This role involves end-to-end ownership of features, from initial scoping with product managers and data scientists through to production traffic and defining success metrics.
Responsibilities
- Ship LLM-powered features and agentic workflows in Python and TypeScript, managing the process from problem definition to production traffic and setting success metrics.
- Develop evaluation metrics, guardrails, and feedback loops to ensure feature trustworthiness and adhere to token and latency budgets.
- Utilize agentic coding tools daily and review their output to identify and correct correctness, security, and design issues before code submission.
- Maintain and enhance NLP pipelines for transforming unstructured claim data into actionable structured data, integrating semantic search, classical NLP, and LLMs.
- Build and operate event-driven services where agents function as steps within deterministic workflows, incorporating checkpointing for model outputs to prevent redundant LLM calls on retries.
- Implement tracing for all model calls and agent steps to enable reconstruction and auditing of recommendations, including citations to source documents and integration with alerts.
- Model and query claims data across relational, document, and graph databases.
- Incorporate privacy and security by design, including tenant isolation, least-privilege access for agents and tools, and clear data handling policies for model providers.
- Author technical design documents for significant changes to gather team feedback and buy-in before development.
- Participate in a shared on-call rotation for built services.
Requirements
- Possess 3+ years of professional software engineering experience, including at least one LLM-powered feature shipped and operated in production.
- Demonstrate proficiency in Python and TypeScript, with typed, tested, async code, and deep expertise in at least one language.
- Have designed LLM output evaluations and shipped guardrails.
- Provide a clear account of how you constrain agentic coding tool context and verify output, based on daily use of tools like Claude Code, Cursor, or GitHub Copilot on production work.
- Experience with pipelines that blend LLMs with semantic search and classical NLP, including structured output, tool calling, embedding-based retrieval, and scenarios where classical techniques outperformed LLMs.
- Experience building services on message- or event-driven architectures, including long-running workflows with retries, idempotent handlers, and event-sourced state, with an LLM or agent step integrated.
- Experience with data modeling and development across relational, document, and graph databases (e.g., PostgreSQL, MongoDB, Neo4j), including appropriate use cases and design decisions for partitioning, consistency, and tenant isolation.
- Experience building and operating backend services on Azure or AWS, with exposure to durable workflow engines like Temporal or Durable Task Scheduler.
- Experience writing technical design documents that have successfully guided team development.
- Exhibit clear written and verbal communication skills for both technical and non-technical audiences, and be capable of giving and receiving code reviews.
Benefits
- Bonus eligible
