COBOL/Mainframe Engineer

Glover Labs is building the agentic platform for legacy software modernization, working with large enterprises whose core business logic runs on mainframe systems. The role involves leveraging mainframe expertise to develop systems that analyze and modernize legacy codebases, collaborating with AI/ML engineers to enhance understanding of business logic and data dependencies.

Responsibilities:

• Build the systems that ingest, parse, analyze, and extract specifications from real mainframe codebases — COBOL, RPG, PL/I, JCL, copybooks, and more
• Apply your deep knowledge of mainframe architecture — batch processing, CICS transaction flows, IMS and DB2 data stores, JCL job streams, VSAM files — to ensure Glover's analysis is accurate, complete, and trustworthy
• Design the domain models and heuristics that allow Glover's AI agents to correctly interpret business logic, data dependencies, and control flow in legacy code — the kind of understanding that only comes from years of hands-on experience
• Collaborate with our AI/ML engineers to improve how large language models reason about legacy code, bridging the gap between what the models can do and what the code actually means
• Work with real customer codebases from Fortune 1000 enterprises and government agencies, validating and refining Glover's output against production-grade mainframe environments
• Ship daily in a modern development environment (Python, TypeScript, cloud-native infrastructure) while bringing your mainframe expertise to the hardest problems in the space

Requirements:

• 5+ years hands-on with mainframe systems — writing, maintaining, debugging, or modernizing COBOL, PL/I, RPG, or Assembler in production. Experience at a major bank, insurer, or government agency is ideal
• Deep working knowledge of the mainframe ecosystem: z/OS, JCL, CICS, IMS, DB2, VSAM, MQ, batch scheduling, and surrounding toolchains (Endevor, ChangeMan, CA7, etc.)
• You understand how business logic is encoded in these systems — not just syntax, but patterns: copybook structures, paragraph naming conventions, batch-online interactions, data flow across programs and JCL steps
• Comfortable (or eager) to work in modern languages and tools alongside your mainframe expertise — Python or TypeScript, Git, CI/CD, cloud infrastructure
• Excited — not reluctant — about applying AI to legacy systems. You see LLMs and agentic AI as tools that, paired with real domain expertise, can finally solve problems brute-force approaches never could
• You want to move fast. You're tired of the pace at large enterprises and ready for an environment where decisions happen in hours and your expertise is the most valuable thing in the room
• Great agency. You see problems, you solve them. You're aligned with a team that believes in using every available tool to multiply its impact