About the role

Software Engineer building and deploying LLM-powered agents for Coalition, an active cyber-insurance and cybersecurity provider. Automating internal business workflows through MCP, evals, and production AI systems.

Responsibilities

Use AI to solve and automate manual work across Coalition
Author reusable skills that agents can invoke
Wire agents to systems of record through MCP servers
Compose skills and tools into working workflows and add MCP surface area when needed
Define correctness with business teams, build eval sets, establish baselines, and determine what ships
Deploy agents from prototype to production with permissions, human-in-the-loop checkpoints, audit trails, and rollback
Tune agents based on real-user behavior and provide documentation and a business-team owner
Partner with data engineers on data access, models, and data quality
Work directly with Coalition teams whose work is being automated and turn processes into a playbook
Individual contributor role with no direct reports

Substantial hands-on experience shipping LLM-backed applications, including prompting, tool use, agentic loops, and multi-step workflows
Shipped at least one agent that non-engineers depend on
Practical experience with MCP, skill or tool authoring, and orchestrating agents against real systems of record
Experience building eval sets and using them to make decisions
Fluent with agentic coding tools like Cursor
Strong Python
Comfortable with SQL and working directly against production data models
Ability to map non-technical operators' processes and translate them into requirements
Git, CI, containers, and operational experience deploying and maintaining work
At least 5+ years of hands-on software development experience
Bonus: experience building or maintaining MCP servers; solutions engineering, professional services, internal tools, or startup founding experience; Linear, Jira, Zendesk, Salesforce, or Snowflake; insurance, fintech, or another regulated domain