Jiffy.com is seeking a Software Development Engineer with deep hands-on experience orchestrating AI coding agents to ship production software at pace. Reporting directly to an Engineering Manager and working closely with Product, GM's and other cross functional teams, you will lead the charge on delivering end-to-end features across our e-commerce platform from storefront and checkout to internal tools and platform services while running multiple AI-driven work streams in parallel.
At Jiffy, engineers no longer write code by hand: every line that reaches production is produced, tested, reviewed, and deployed by coding agents (Claude Code, Codex, Cursor) under the direction of a human orchestrator. This role is perfect for an individual contributor who has already internalized the shift from “typing code” to “directing agents,” has strong architectural taste and review instincts, and wants to help define what an AI-native engineering org looks like. You will thrive here if you are a systems thinker, a builder by nature, and someone who gets more done by managing five parallel agents well than by writing a single function perfectly.
At this time, Jiffy is not able to sponsor or take over sponsorship of an employment visa (e.g., H-1B) for this position. Applicants must be legally authorized to work in the United States without employer sponsorship, now or in the future.
Key Responsibilities
- Orchestrate AI coding agents (Claude Code, Codex, Cursor) across 3–6 concurrent work streams spanning features, refactors, bug fixes, and investigations using worktrees, sub-agents, and background tasks to maximize parallel throughput.
- Own features end-to-end: translate product intent into clear specifications, decompose work into agent-executable plans, drive agents through implementation, review every diff critically, run QA, land PRs, and verify production health.
- Design and architect scalable, reliable systems across our stack Rails 8 / Solidus e-commerce, GraphQL APIs, React 19 front end, Kubernetes infrastructure holding the architectural and judgment calls that agents cannot make.
- Raise the quality bar on AI-generated code by catching hallucinated APIs, enforcing secure and idiomatic patterns, and pushing back on output that is technically working but structurally wrong.
- Build and evolve Jiffy’s internal AI tooling: author skills, rules, prompts, MCP integrations, and evaluation harnesses that make the whole engineering org faster. Contribute to our shared .ai/ skills and rules library.
- Operate with production rigor: drive agents to produce meaningful test coverage (RSpec, Vitest, Playwright), monitor deploys, respond to incidents, and fully own the services you ship.
- Partner cross-functionally with Product, Design, Ops, and peer engineering teams to define and ship features; participate in architecture and code reviews; mentor engineers on AI-native workflows and orchestration patterns.
- Continuously identify where remaining manual work (QA, ops runbooks, data investigations, migrations) can be replaced by agents, and automate it.
