Rapid7

Principal AI Software Engineer, Agentic Platform

Rapid7
Apply
23 hours ago
Belfast, United KingdomStaff+

Responsibilities

  • Evolve the architecture of the agentic loop from change detection through evidence-based risk verification.
  • Design and build remediation reasoning for fixes, mitigations, dependency-aware planning, customer policies, approvals, and change windows.
  • Develop safe, reversible autonomous actions with approval, execution, verification, rollback, and trusted audit trails.
  • Build agent evaluation, golden datasets, regression suites, LLM-as-judge workflows, and online quality metrics.
  • Improve agent context using exploitability, threat intelligence, reachability, attack paths, asset criticality, and ownership.
  • Improve agent latency, cost, and reliability through model selection and routing, token management, caching, and provider failover.
  • Write production code, review designs, pair with engineers, and raise technical standards across teams.
  • Partner with Product and Engineering leadership to define priorities and connect technical decisions to customer and business outcomes.

Requirements

  • 10+ years of software engineering experience, including significant senior technical contribution on complex, high-scale distributed systems.
  • Recent hands-on production experience building LLM-based agent systems involving tool use, planning, retrieval, context engineering, and evaluation.
  • Strong understanding of autonomous-system trade-offs, including control, reversibility, latency, reasoning depth, model capability, and cost.
  • Hands-on experience with foundation model platforms, LLM gateways or routers, agent frameworks or protocols, LLM evaluation tooling, durable workflow engines, and event streaming.
  • Production-quality programming ability in Python or Go.
  • Deep understanding of distributed systems and designing for failure at scale in cloud-native environments, preferably AWS and Kubernetes.
  • Experience with vulnerability or exposure management, patch and configuration management, or security orchestration and automated response is preferred.
  • Experience with agent AI safety practices such as prompt-injection defenses, sandboxed execution, and least-privilege tool access is preferred.
  • Ability to own significant technical outcomes across multiple teams, resolve complex challenges, communicate technical trade-offs, and build cross-functional partnerships.
  • Ability to connect customer needs and business priorities to technical decisions and evaluate the value, cost, and trade-offs of AI solutions.
Rapid7

About Rapid7

5,001-10,000 employees

Rapid7 builds a cloud security operations platform and services for enterprises, combining vulnerability management, detection and response (SIEM/XDR and MDR), application security testing, and security automation. Its Insight platform products include InsightVM, InsightIDR, InsightAppSec, and InsightConnect, sold as subscriptions with optional managed services. Founded in 2000 and headquartered in Boston, the NASDAQ-listed company (RPD) serves more than 11,500 customers worldwide.

Contact me