AI Engineer

Fractal Analytics is seeking a hands-on AI Engineer to build and ship production-grade software using coding agents. The role focuses on configuring and integrating those agents so they operate reliably across enterprise codebases, alongside automation, security guardrails, evaluation, and internal enablement for Financial Services transformation.

• Use coding agents (Claude Code, Codex, Cursor, Git Hub Copilot, and similar) daily to deliver production-quality code including features, refactors, migrations, tests, and documentation.
• Apply agents to legacy modernization, code migration, and platform migration across large, multi-repo enterprise environments.
• Act as a reference implementer by proving patterns on real projects and codifying them for broader team adoption.
• Review, test, and harden agent-generated changes to meet production standards.
• Build and maintain standardized agent configurations, including project memory files (CLAUDE.md, AGENTS.md, Cursor rules), custom slash commands, skills, subagents, and hooks.
• Develop MCP servers and tool integrations connecting agents to internal systems such as repositories, ticketing, CI/CD, documentation, and internal APIs.
• Engineer context packages, prompt templates, and codebase indexing approaches for large and legacy codebases.
• Tune agent behavior and diagnose failure modes such as context overflow, tool misuse, and looping.
• Build headless and background workflows using CLIs and SDKs, including Claude Code headless mode, Claude Agent SDK, and Codex CLI.
• Implement agent automation within CI/CD pipelines, including automated PR review, test generation, dependency upgrades, vulnerability remediation, and migration jobs.
• Create internal tooling and wrappers to make agents safer and easier to run across repositories and teams.
• Automate measurement and evaluation of agent output quality through tests and regression checks.
• Implement technical controls for safe agent usage, including permission models, sandboxing, command allow lists, secrets handling, and data exposure protections.
• Integrate agents with governed Claude access using AWS Bedrock, IAM, and the enterprise AI Gateway.
• Build audit logging and usage tracking for agent activity and AI-generated code.
• Partner with Info Sec, Risk, and Compliance to translate Financial Services regulatory and model risk requirements into enforceable technical controls.
• Develop telemetry pipelines and dashboards for usage, token consumption, cost, and quality metrics such as acceptance rates, cycle time, and defect rates.
• Run hands-on bake-offs of emerging agentic coding tools and models against real enterprise tasks, producing data-driven recommendations.
• Document reference implementations and pair with engineering teams to unblock adoption.
• Contribute reusable configurations, templates, and examples to a shared internal repository.

• Bachelor's or Master's degree in Computer Science, Engineering, or a related discipline (or equivalent experience).
• 5+ years of professional software engineering experience, with strong proficiency in Python and at least one additional language (Type Script or Java).
• Proven, daily hands-on use of Claude Code, Codex, and Cursor (or similar agents) delivering production code, including examples of non-trivial shipped work.
• Experience building MCP servers and/or custom tools such as skills, subagents, or hooks.
• Strong understanding of context and prompt design, task decomposition, and methods to steer and verify agent output.
• Experience building CI/CD automation and scripting agent workflows via CLIs or SDKs.
• Knowledge of Git workflows, testing, code review, and secure coding practices.
• Working knowledge of AWS services including Bedrock, IAM, Lambda, API…