F

FICO

Predictive credit risk scoring & analytics

Principal Harness Engineer

Software EngineeringConfirmed live in the last 24 hours
Full-timeRemote in UKNo salary listedRemote

About the Job

Requirements

  • A Bachelor's or Master's degree in Computer Science or a related discipline, or relevant commercial experience in software architecture, design, development, and testing.
  • Seasoned software engineering experience with large, complex codebases and a strong foundation in architecture, design, testing, and maintainability.
  • Hands-on experience with AI coding agents such as Claude Code, Codex, or similar tools.
  • Experience building engineering tooling across a modern stack, including linters, static analysis, continuous integration and continuous delivery pipelines, containerised build and test environments, instrumentation, and observability.
  • Experience with spec-driven development, context engineering, agent orchestration, fitness functions, and developer-platform work.
  • Ability to encode engineering quality standards into mechanical, repeatable rules.
  • Understanding of the cost, speed, and reliability trade-offs between deterministic computational controls and inferential large-language-model-based controls.
  • Experience owning quality-gating processes and defining release criteria to ensure engineering standards are consistently met.
  • Working knowledge of autonomous-agent security, including prompt injection, tool and permission scoping, sandboxed execution, audit trails for agent actions, and least-privilege guardrails.
  • Experience with consumer and contract testing approaches such as Pact to validate service integrations across distributed systems.
  • Ability to articulate designs with architects and drive standards across teams.

Responsibilities

  • Design, build, deploy, and support core harness components, including guides, feedback loops, guardrails, and shared context that turn model capability into production-grade engineering.
  • Own and evolve feedforward guides, including agent instruction files, reusable skills, architectural rules, reference documents, and codemods, and drive team-wide standardisation.
  • Build feedback sensors, including custom linters, static analysis, structural and architecture-fitness tests, verification loops, and large-language-model-as-judge reviewers.
  • Own quality gating and release criteria for agent-produced work, including authority boundaries and escalation rules for human review.
  • Establish large-language-model testing infrastructure and evaluation approaches for AI-generated output quality and safety, applying consumer and contract testing where service integration reliability matters.
  • Run the steering loop by engineering controls that prevent repeated agent mistakes, and maintain repository knowledge as the system of record through continuous garbage collection.
  • Decide where controls run in the path to production, including pre-commit checks, post-integration checks, and continuous drift sensors.
  • Improve observability into agent work and track measures such as cost per merged pull request, merge time, review velocity, defect escape rate, and agent-pull-request survival rate.
  • Partner with product and platform teams to turn specifications and acceptance criteria into enforceable controls.
  • Serve as a source of technical expertise and mentor engineers across teams in harness practices and responsible use of AI tools.

About the Company

FICO

Predictive credit risk scoring & analytics

FICO provides data analytics and decision-management tools for credit risk and lending, with the FICO score as its flagship measure of consumer credit risk used by lenders. The score is calculated from a user’s credit report data to predict loan repayment likelihood, and FICO software applies analytics and decision rules to guide loan approvals, pricing, fraud detection, and account management. Unlike many competitors, FICO combines a widely adopted score with an end-to-end suite of predictive analytics and decision-management tools that cover the full lending lifecycle. Its goal is to help financial institutions make better lending and risk decisions, prevent fraud, and optimize customer relationships through data-driven, rules-based systems.

Company size
1,001-5,000
Company stage
IPO
Headquarters
San Jose, CA
Founded
1956

Recently Posted Jobs

Explore more opportunities at FICO

See more jobs