Description
Software delivery is still stuck in slow motion. Backlogs grow, cycle times are measured in weeks, and “code review” too often means rubber-stamping to hit a sprint line. Most teams are drowning in integration risk, brittle architectures, and unclear requirements. Meanwhile, AI coding agents are shipping credible code, but only when they’re fed precise specs and supervised with rigor. The industry keeps throwing more humans at the keyboard instead of fixing the system that turns intent into reliable, production-grade software.
Trilogy is taking the opposite path. We run an agentic development workflow end-to-end: engineers drive research, specification, architecture, and quality, while AI agents handle implementation. Specs are written to be executable by Claude Code, test-first and surgically scoped to minimize blast radius. The result is continuous delivery of production features with dramatically tighter feedback loops, fewer surprises in integration, and higher confidence in correctness before a single line lands on main.
This job is for the engineer who wants to design systems and verify correctness, not grind tickets. You’ll spend roughly 10–20% in research and technical discovery to understand existing behavior, constraints, and stakeholder needs; 40–50% converting that understanding into precise, test-driven, agent-readable specifications and architectural decisions; and 40–50% orchestrating and supervising implementation. You are accountable for what ships: no silent failures, no ambiguous specs, no “the agent did it.”
You’ll join a team that treats specification as a product, review as an engineering discipline, and delivery as the only KPI that counts. If you’re energized by elegant system design, surgical scoping, and ruthless validation—and you already use tools like Claude Code, Cursor, Copilot, or ChatGPT as part of your daily workflow—you’ll thrive here. Bring your architecture chops, your TDD instincts, and your bar-raising taste in code quality. If this sounds like you, step in and lead the future of how software gets built. Apply and show us how you orchestrate agents to deliver production-grade outcomes at scale.
What you will be doing
- Technical Discovery: Investigating existing systems, stakeholder requirements, and data sets to understand the problem space and technical constraints
- Technical Specification: Converting research findings and stakeholder requirements into precise, executable technical specifications designed for agentic coding assistants (primarily Claude Code) using spec-driven development methodology
- AI Implementation: Orchestrating AI coding agents to implement specifications, continuous review, and validation of agent-produced code, and integrating validated changes into the production codebase
Key responsibilities
- Continuous delivery of production-grade software features and systems using agentic development workflows, where AI coding agents handle implementation while engineers drive specification, architecture, and quality assurance
Candidate requirements
- At least 3 years professional software engineering experience in production environments
- Experience participating in pull request review processes
- Experience using AI coding tools (e.g. Claude Code, Windsurf, Cursor) in a professional setting
Applying for a role? Here’s what to expect.
Crossover's skill assessment process combines innovative AI power with decades of human research, to take the guesswork, human bias, and pointless filters out of recruiting high-performing teams.
Chat-style screening interview.
Cognitive aptitude test.
Prove real-world job skills.
Interview with the hiring manager.
Accept Job Offer.
Pass proctored test.
Frequently asked questions
Reviews
Meet some people who've landed similar jobs
Recruitment sucks. So we’re fixing it.
The Olympics of work
It’s super hard to qualify—extreme quality standards ensure every single team member is at the top of their game.
Premium pay for premium talent
Over 50% of new hires double or triple their previous pay. Why? Because that’s what the best person in the world is worth.
Shortlist by skills, not bias
We don’t care where you went to school, what color your hair is, or whether we can pronounce your name. Just prove you’ve got the skills.
Join the world's largest community of 
AI-first remote workers.