AI Workflow Engineer

Accertify, Inc.•Itasca, IL

1d•$115,000 - $120,000•Remote

About The Position

Accertify enables commerce by doing one thing extraordinarily well: pinpointing fraud. The company's Predictive Yes Platform helps businesses say yes to more - more good customers, more revenue, and more growth - without getting burned. It's a precise, confident, data-backed yes, made possible by joining signals across the entire customer lifecycle through unmatched data volume, layered AI-powered models, and a consortium approach that protects clients from bad actors spotted anywhere in the network while unlocking business with good actors verified anywhere in the network. With more than 10 billion transactions and over $1 trillion in commerce processed in 2025 alone, Accertify delivers the intelligence and precision that fraud and payments teams need to say yes confidently and enable growth. That's the More Yes difference. We're looking for a Workflow Engineer who turns AI capabilities into production-grade agent workflows that automate real business processes. You'll design and ship LLM-powered flows, from RAG pipelines and MCP tools to multi-step agentic automations, and integrate them safely with our apps, data, and APIs. You'll work closely with product, platform, and security to deliver fast, reliable, and cost-efficient AI outcomes.

Requirements

Strong coding in Python and/or Node; clean API design
Hands-on with LLM integration (prompting, tool use, function calling)
Experience building RAG systems and working with vector databases.
Practical familiarity with LangGraph/CrewAI/n8n or similar orchestration.
Experience building MCP tools/servers or equivalent agent-tool frameworks
Solid understanding of REST/GraphQL, auth (OAuth/OIDC), and secure integration.
Containers, Git, CI/CD basics; comfort shipping production services.

Nice To Haves

Evals & guardrails (e.g., structured output, JSON schemas, toxicity/PII checks)
Knowledge graphs / event-driven architectures/workflow engines
Model selection & optimization; prompt caching; cost/perf tuning
Exposure to LLM safety, privacy, governance, and enterprise compliance

Responsibilities

Build agentic systems: Design and implement agent workflows using tools like LangGraph, CrewAI, n8n (or similar); Create task decomposition, tool-use policies, guardrails, and evaluation loops
Implement Retrieval-Augmented Generation (RAG): Stand up vector search and retrieval pipelines; select embedding strategies; Optimize context packing, grounding quality, and hallucination mitigation
Develop MCP tools & integrations: Build Model Context Protocol (MCP) servers and custom tools to safely connect to APIs, databases, and internal systems; Establish capabilities, permissions, and auditing for tool use
Ship production-ready AI services: Build FastAPI/Node services that expose AI workflows behind secure APIs; Own CI/CD for workflow services; write tests, contract checks, and evals.
Measure and improve: Instrument workflows with latency, cost, accuracy, and safety metrics; Run prompt/eval experiments; iterate on prompts, memory, and grounding
Partner across teams: Work with product & ops to prioritize use cases; Collaborate with Platform/DevOps to deploy and scale reliably in Kubernetes