Agentic Systems Engineer

We are looking for a Senior Engineer with deep agentic development experience and Java/Kotlin expertise to design and build the agentic backend layer powering Unlimit’s global financial platform.

You will not just write backend services using your harness. You will design AI agents that generate, operate, validate, monitor, and evolve backend services.

This role is for engineers who understand that the future of software is Human + AI co-creation, and soon — AI-to-AI collaboration.

What You’ll Be Doing

• At Unlimit, agentic engineering is how we build software. AI agents are the primary execution layer of our engineering organization, while engineers provide the architecture, direction, constraints, and judgment that make autonomous development possible.

• You will orchestrate AI agents that design, implement, test, refactor, and maintain production-grade Java and Kotlin systems. Manual coding is not part of our engineering model – software implementation is performed by AI agents. Your deep expertise in Java, Kotlin, and backend engineering remains essential: you must understand architectural trade-offs, evaluate generated implementations, identify subtle defects, and ensure production-grade performance, security, and reliability. You will translate complex engineering challenges into agent-executable work, establish clear specifications and acceptance criteria, and coordinate parallel development workflows. You own the engineering decisions and outcomes, while agents execute the implementation.

• You will engineer the environment in which agents operate. This means building agent-native development workflows, maintaining machine-readable architecture and documentation, defining reusable skills and tools, and creating the context infrastructure that enables agents to navigate large codebases, make informed decisions, and execute complex tasks autonomously. Repositories, APIs, development environments, and engineering processes must be designed for machines as first-class participants.

• You will establish continuous, automated verification across the entire software lifecycle. AI-generated implementations must be validated through deterministic tests, architectural constraints, security controls, and regulatory requirements, including PCI DSS and financial compliance standards. You will design feedback loops where agents detect failures, investigate causes, correct implementations, and independently verify results, with appropriate human oversight for high-risk decisions.

• Your work will extend beyond development into agent-driven operations and production reliability. You will build systems that enable agents to understand runtime behavior, investigate incidents, identify performance bottlenecks, and propose or execute controlled remediation. Observability, operational tooling, and infrastructure will become interfaces through which intelligent systems safely interact with our global payment infrastructure.

• You will collaborate across engineering, platform, security, and data teams to build a coordinated ecosystem of specialized agents, sharing capabilities, context, and execution standards. You will continuously challenge traditional engineering practices, eliminate unnecessary human coordination and repetitive work, and push the boundaries of what a small team of exceptional engineers can deliver.

Qualifications You Will Bring

• An agent-native engineering mindset. You see AI agents as the primary execution layer of software engineering, not coding assistants. You are comfortable working in an environment where manual coding is prohibited, and your value comes from architecture, specifications, orchestration, verification, and engineering judgment.

• Proven experience with autonomous agentic development. You have built production-oriented engineering workflows using Claude Code, OpenAI Codex, or comparable platforms, leveraging parallel agents, subagents, reusable skills, hooks, MCP servers, and automated execution pipelines.

• Mastery of specification-driven development (SDD). Practical experience with methodologies and frameworks such as GitHub Spec Kit, BMAD Method, or equivalent approaches that transform business requirements into structured specifications, executable plans, and independently verifiable agent tasks. You understand how to maintain alignment between requirements, architecture, implementation, and tests.

• Strong context engineering capabilities. You know how to structure repositories, engineering knowledge, agent instructions, and persistent context using mechanisms such as AGENTS.md, CLAUDE.md, reusable skills, and machine-readable documentation. You understand context-window management, progressive disclosure, retrieval strategies, context compaction, and long-running agent execution.

• Experience orchestrating multi-agent engineering workflows. You can decompose complex initiatives into parallel workstreams, coordinate specialized agents, manage dependencies and isolated execution environments, and establish reliable integration and verification processes. You understand when agent autonomy creates leverage and when additional orchestration introduces unnecessary complexity.

• Hands-on experience with frontier reasoning and coding models. You understand the capabilities and limitations of Claude, GPT/Codex, and comparable models, including model selection, reasoning effort, tool use, structured outputs, token economics, and cost-performance trade-offs across different engineering tasks.

• Expertise in agentic quality assurance and autonomous verification. You design deterministic quality gates combining automated testing, static analysis, architectural fitness functions, security scanning, and independent agent reviews. You understand test-driven development, adversarial validation, and iterative agent feedback loops, ensuring that generated software meets measurable acceptance criteria.

• Strong Java or Kotlin engineering expertise. At least 4 years of production experience with Java (Spring/Spring Boot) or Kotlin, including concurrency, performance optimization, database interactions, and application architecture. You can critically evaluate agent-generated implementations, recognize subtle defects, and enforce engineering standards without manually writing application code.

• Deep knowledge of distributed systems and financial-grade backend engineering. You understand scalable microservices, event-driven architectures, transactional consistency, fault tolerance, observability, and high-throughput, low-latency processing. Experience with payment systems, PCI DSS, and regulated financial infrastructure is a strong advantage.

• A security-first approach to autonomous engineering. You understand agent permissions, sandboxed execution, secrets management, supply-chain security, prompt injection risks, policy-as-code, and controlled deployment workflows. You can establish clear boundaries between autonomous execution and actions requiring human authorization.

• An engineering productivity and continuous improvement mindset. You measure agentic development by delivery quality, lead time, reliability, cost, and engineering throughput rather than generated code volume. You continuously experiment with emerging methodologies, evaluate new capabilities, and improve the autonomous engineering workflows used by your team.

• Strong communication and technical leadership skills. You can articulate complex architectural decisions, challenge assumptions, collaborate across engineering, product, security, and compliance, and take end-to-end ownership of outcomes in a globally distributed AI-native organization.

Nice to have:

• Experience in fintech, payments, or banking systems.

• Experience working with AWS services and infrastructure.

• Familiarity with containerized deployments (Docker, Kubernetes) and CI/CD pipelines.

• Exposure to AI-enabled systems or interest in integrating intelligence into applications.

• Background in event-driven architectures and message-based systems (Kafka, RabbitMQ).

Why This Role Exists:

At Unlimit, autonomous engineering is our operating model. AI agents execute the software development lifecycle, while engineers architect, orchestrate, and own the outcomes.

We are building a global financial platform designed for both humans and autonomous agents.

If you want to:

Engineer autonomous systems that build and evolve software

Push the limits of specification-driven, multi-agent engineering

Build financial infrastructure where AI agents are first-class participants

Shape the next generation of AI-native fintech at global scale

Then this is your role.