AI-Native Software Engineer

About This Opportunity

We are representing a confidential SaaS organization that is moving beyond simple AI autocomplete to a fully agent-driven engineering workflow. This is not a role where you build LLM products—this is a role where you build production features with agents doing most of the writing while you set intent and own outcomes.

You'll work inside a real, high-traffic production codebase processing billions of requests daily. You'll plan, implement, test, review, debug, and ship features using coding agents, and you'll help build the tooling that makes this workflow scalable for the entire engineering team.

If your first reaction to a repetitive workflow is "this should be an agent," we should talk.

Responsibilities

• Ship production features with agents: Use Claude Code, Codex, Cursor, Copilot, or similar tools across the full development lifecycle—planning, implementation, testing, debugging, refactoring, documentation, and review. Maintain the same quality bar as hand-written code.
• Own the output: Review agent-generated code as you would review a junior engineer's work. Watch for hallucinations, wrong assumptions, hardcoded secrets, missing validation, and injection-prone patterns. For authentication, authorization, cryptography, and session handling, use standard implementations rather than model inventions.
• Maintain context quality: Keep CLAUDE.md, coding standards, architecture notes, and module dependencies current so agents succeed on the first attempt. Feed SAST findings back with call path and data flow for real patches, not suppressions.
• Build team-level tooling: Create prompts, skills, sub-agents, and MCP integrations for recurring work. Establish guardrails that keep agents within architectural limits and away from forbidden patterns. Monitor token costs.
• Use agents beyond code generation: Apply AI to integration tests, coverage gap analysis, regression tests from production incidents, root-cause analysis with logs and traces, and draft ADRs and runbooks.
• Review AI-assisted work effectively: Run context-aware AI review on your own diffs, distinguish true from false positives, and convert patterns into team rules.
• Share effective practices: Demonstrate your setup to teammates, review AI-assisted pull requests, and get at least one of your tools adopted beyond your own repository.

What We're Looking For

Must have:

• Daily use of coding agents on real production work—not courses or demos. You must have shipped code where agents did substantial writing and you owned the result.
• At least one project you can walk through end-to-end: what you delegated, what you kept, where agents failed, and how you caught errors.
• Agentic experience including tool and function calling, multi-step workflows, state management, error recovery, and human-in-the-loop patterns.
• Context discipline: demonstrated ability to set agents up for success through project files, standards, examples, and scoped tasks that reduce rework.
• Solid engineering fundamentals: clean code, testing, error handling, security, and performance awareness. Agents amplify weak fundamentals.
• Judgment to recognize when agents are wrong. We test for judgment, not just speed.
• Willingness to share knowledge. You show teammates your setup rather than hoarding practices.

Language is not a primary filter. Go and PHP/Laravel are preferred for our stack; Python, TypeScript, or other languages are acceptable if fundamentals are strong.

Nice to have:

• Team-level tooling you've built and others have adopted: shared skills, sub-agents, MCP servers, slash commands, or review/test agents in CI.
• Security habits around agents: forbidden patterns, least-privilege tool access, secrets management, and verification of AI-suggested packages before merge.
• Testing and debugging depth with AI: coverage gap analysis, regression tests from incidents, root-cause work using observability data.
• LLM work: prompt engineering, agent SDKs, or orchestration frameworks like Strands Agents, LangChain, or LlamaIndex.
• Agentic patterns beyond basics: RAG, MCP, skills, hooks, plugins.
• Experience designing scalable, highly available systems, ideally in Go.
• Curiosity about system internals: benchmarking, breaking, and improving these tools.

On seniority: We evaluate mid-level and senior candidates. Your level comes from what you've shipped, not your title. The non-negotiable: agents must already be how you work daily.

How We Work With You

Candidates apply through Get A Job.ai. Our talent team will screen your application and conduct an initial conversation about your agent-driven engineering experience. If there's a strong match, we'll submit your profile to our client for their review process. Please do not contact the client directly—all communication flows through our recruiting team to ensure a smooth, professional process.

Pay

Compensation details will be discussed during the screening process based on experience level and our client's structure.

Equal Opportunity: Get A Job.ai provides equal opportunity in a zero-discrimination process. We welcome and embrace everyone without regard to sex, race, color, nationality, religion, gender identity, sexual orientation, disability status, citizenship, or marital status.