AI Refactoring Manager

Hive Financial Systems•Atlanta, GA

12d•Onsite

About The Position

Hive Financial Systems is hiring an AI Refactoring Manager to lead the rebuild of its legacy lending platform into a multi-portfolio architecture. The role will establish and operate an AI-assisted engineering workflow in which AI coding agents generate much of the code while people own technical judgment, review, and quality. This is a hands-on leadership position that includes setting the method, running the work, and reviewing code. The legacy platform comprises roughly 50 repositories built portfolio by portfolio. The target is a microservices architecture using .NET, Azure, and Kubernetes that can support 30 lending portfolios through configuration rather than code forks. This is a rebuild rather than a line-by-line port: legacy code is a source for understanding business needs, not a model for the new architecture. Because the work concerns regulated lending, correctness in financial calculations, disclosures, and borrower-data protection is essential. In the first 12 months, the manager is expected to establish a repeatable AI-assisted rebuild method used daily by every rebuild team; ensure Product signs off on extracted business rules and acceptance criteria before each service rebuild; achieve at least 90% line and branch test coverage for every release, with failing tests written first; improve lead time and deployment frequency without increasing change failure rates; and put agreed legacy capabilities live on the new platform with legacy equivalents switched off, without compliance or data-isolation defects from AI-generated code reaching production.

Requirements

At least 8 years of software engineering experience, including at least 3 years leading engineers or a technical program.
Has led at least one large legacy modernization, such as a monolith-to-microservices migration, re-platform, or rewrite, through to production.
Proven experience creating and operating agentic AI engineering workflows on production work, including multi-step workflows using Claude Code or similar frameworks, and understanding their failure modes.
Strong C#/.NET skills and comfort reading older .NET and SQL Server code.
Practices test-driven development and can explain why coverage figures alone do not establish quality.
Working knowledge of CI/CD, trunk-based development, and cloud deployment; Azure is preferred.

Nice To Haves

Experience in lending, payments, or another regulated financial domain, including TILA/Reg Z, NACHA, FCRA, or GLBA.
Experience with Kubernetes, Helm, Terraform or Bicep, and GitOps tooling such as Argo CD.
Has written agent instructions, prompt libraries, or internal AI tooling used by a team.
Experience designing multi-tenant systems where tenant data isolation is a hard requirement.
Familiarity with DORA metrics and observability tools such as DataDog.

Responsibilities

Own the rebuild roadmap and sequence legacy capabilities based on business risk, dependencies, and value.
Design and run the AI-assisted engineering workflow for reading legacy code, extracting business rules, drafting services, and handing work to human reviewers.
Build and maintain prompts, agent instructions, and reusable skills that encode Hive’s architecture standards.
Enforce test-first engineering: establish failing acceptance tests before code is written and apply the same quality gates to AI-generated and human-written code.
Review AI-generated code and set the team’s review bar, with particular attention to financial calculations, borrower data, and portfolio isolation.
Distinguish live code from dead code in the legacy estate and document embedded business rules before they are lost.
Partner with Compliance and Product to capture and test regulated behavior, including disclosures, payments, adverse action, and consent.
Track and report delivery health metrics, including lead time, deployment frequency, change failure rate, and recovery time, as well as AI effectiveness measures such as rework rate and defects traced to AI output.
Coach engineers on effective use of AI agents and determine where AI should not be used.
Evaluate AI coding tools and models, and manage their cost, access, and security posture.