About this role
Engineers who thrive in zero-to-one ambiguity and enjoy turning messy real-world workflows into robust automated systems will find a natural home in this role. Workstream is expanding its product portfolio by building AI-native services designed to deliver completed business outcomes through autonomous agents backed by human review. As the founding engineer on this new team, you will spearhead development from scratch, shaping the technical architecture and setting engineering standards for future hires.
You will work directly with design customers to understand operational realities and build production-ready agents, orchestration pipelines, evaluation harnesses, and human-review tooling. This position exists to establish a brand new company initiative that automates critical business operations for the hourly workforce. Success requires strong engineering judgment, deep care for output reliability, and the willingness to handle everything from complex data plumbing to unglamorous review interfaces.
What you will do
- Build and operate production-grade AI agents and agentic workflows covering orchestration, tool use, data connections, and deterministic logic.
- Collaborate directly with design customers to map real-world workflows and build solutions against live data.
- Develop comprehensive evaluation harnesses including pre-launch thresholds, regression suites, and post-launch monitoring.
- Design the human-review layer consisting of exception queues, review tooling, escalation paths, and decision-tracing mechanisms.
- Execute shadow-mode parallel runs against existing customer processes to eliminate discrepancies prior to go-live.
- Productize customer-specific implementations into reusable agents, integrations, and playbooks.
- Handle production incidents end to end and integrate failure learnings back into evaluation suites and system controls.
What you need
- Personally built and operated at least one production-grade AI agent or agentic workflow with a clear understanding of failure modes and evaluation methods.
- Demonstrated strong engineering judgment spanning models, orchestration, APIs, data integrations, permissions, and observability.
- Direct experience working with customers or end users through discovery, implementation, or incident resolution.
- Ability to decompose messy operational workflows into deterministic steps, agentic decisions, human judgment points, and exception paths.
- Deep commitment to preventing plausible but incorrect AI output through grounded context, constrained actions, and verification.
- Comfort thriving in zero-to-one ambiguity by shipping weekly and owning outcomes across diverse technical tasks.
Nice to have
- Exposure to a regulated domain such as payroll, payments, tax, benefits, or labor compliance
Benefits
- Competitive salary and equity
- Comprehensive health coverage including medical, dental, and vision with 90% paid for employees and 80% for dependents
- 401K Plan
- Flexible PTO
Skills and tech stack
Questions candidates ask
What is the remote work policy for this role?
This is a full-time, remote position with occasional in-office visits as needed to support close cross-functional collaboration and key team initiatives.
What will I actually own on the team?
You will own the technical components of the AI-native services, including whether they work, how they fail, and how fast bespoke work becomes reusable. Pricing, deal acceptance, and the margin model sit with the Head of Delivery and the Head of GTM.
What kind of experience is most important for this position?
Demonstrated production experience matters more than title or years. You must have personally built and operated at least one AI agent or agentic workflow in production.