AI Workflow Engineer

Vindicor turns a new Stripe dispute into a traceable case with Shopify context, evidence state, a recommended next step, and operator approval. As an AI Workflow Engineer, you will build the orchestration that keeps that process reliable from intake through review.

This role sits at the boundary between integrations, workflow state, model-assisted drafting, deterministic validation, and operational tooling.

• Build stateful workflows for dispute intake, evidence assembly, review, and outcome tracking
• Integrate APIs, webhooks, queues, retries, and idempotent background jobs
• Design model-assisted steps with explicit inputs, evaluations, and deterministic safeguards
• Make missing data, conflicts, and failed transitions visible to operators
• Add observability and audit records across long-running workflows
• Test recovery paths so a partial failure does not corrupt the case history

• Build stateful workflows for dispute intake, evidence assembly, review, and outcome tracking
• Integrate APIs, webhooks, queues, retries, and idempotent background jobs
• Design model-assisted steps with explicit inputs, evaluations, and deterministic safeguards
• Make missing data, conflicts, and failed transitions visible to operators
• Add observability and audit records across long-running workflows
• Test recovery paths so a partial failure does not corrupt the case history

You will own workflows where correctness is visible: inputs have sources, transitions have states, and operator decisions remain in the record. The role works directly with product and backend engineering.

Compensation, equity, schedule, time off, equipment, and other employment terms are discussed directly during the interview process.