Senior Integrations & AI Workflow Engineer

Role Overview

We are seeking a Senior Integrations & AI Workflow Engineer (Contractor) to own two connected halves of our platform: the integrations that connect ours to the systems our partners run on, and the AI workflows that remove manual effort from how we operate and build.

This is a hands-on technical role with cross-team scope. You will:

● Design and own the integration architecture between ours and external LMS, identity, content, and employment partners.

● Build AI and agentic workflows that automate internal operations, from support triage to data reconciliation to facility onboarding.

● Raise engineering throughput by bringing AI tooling into the development lifecycle in ways the team actually adopts.

● Establish the evaluation, guardrail, and observability practices that make AI output trustworthy in a corrections environment.

● Set technical direction across teams for both integrations and AI workflows, and bring other engineers along.

The right candidate is equally comfortable debugging an OAuth handshake with a vendor and deciding whether a workflow needs an agent, a deterministic pipeline, or neither. You should be skeptical of AI for its own sake and willing to show a measured benefit before a workflow ships.

Key Responsibilities

Partner & Platform Integrations

● Own the design, build, and lifecycle of integrations with learning management systems, identity and access providers, content libraries, and employment partners.

● Define the integration architecture and standards other teams build against: contracts, versioning, idempotency, retry and backfill strategy, and failure semantics.

● Build resilient sync pipelines for rostering, enrollment, progress, and completion data, with reconciliation you can run and trust.

● Lead technical discovery with vendor engineering teams, including API review, sandbox validation, and security questionnaires.

● Instrument integrations so that a partner-side outage is visible to us before it is visible to a facility.

AI Workflow Engineering (Internal Operations)

● Identify recurring, high-volume manual work across engineering, support, and operations, then design AI workflows that remove it.

● Build LLM-backed and agentic automations for support triage, data reconciliation, reporting, and facility onboarding.

● Choose the right tool for each problem, whether that is deterministic code, retrieval, tool-calling agents, or human review in the loop, and document the reasoning.

● Integrate workflows with the systems they act on (ticketing, databases, cloud APIs, internal services) using durable orchestration rather than brittle scripts.

● Measure the outcome of every automation in hours saved or errors avoided, and retire the ones that do not pay for themselves.

Engineering Productivity & Agentic Tooling

● Build and maintain agentic tooling for the engineering org: coding agents, test generation, review automation, and CI enrichment.

● Create the shared context that makes those tools effective, including repository conventions, tool definitions, and MCP servers over internal systems.

● Drive adoption through documentation, working sessions, and templates, and measure it by real usage rather than by licenses issued.

Evaluation, Guardrails & Observability

● Stand up the evaluation infrastructure for AI workflows: golden datasets, regression runs on every prompt or model change, and quality thresholds that gate merges.

● Implement guardrails for unsafe or invalid input and output, with escalation paths for anything touching learner data or facility policy.

● Version prompts, models, and retrieval configuration alongside code so behavior changes are reviewable.

● Build tracing and cost observability for AI workflows, and set the rules for when a human stays in the loop.

Security & Compliance

● Apply secure-by-default practices to every integration and workflow: secrets management, least-privilege access, and data minimization.

● Ensure AI workflows and partner integrations meet SOC2 expectations, and support audit preparation with automated evidence.

● Treat PII and justice-impacted learner data with the handling controls our environment requires, including decisions about what never leaves our boundary.

Must-Have Skills

● Proven experience at a senior or staff level building production integrations and backend services.

● Deep API engineering background: REST, webhooks, OAuth 2.0 and OIDC, SAML, and asynchronous, event-driven patterns.

● Strong experience with third-party and vendor integrations, including owning the relationship with partner engineering teams.

● Production experience building LLM-backed systems: prompt engineering, retrieval, tool calling, and agent orchestration.

● Demonstrated ownership of AI evaluation and guardrails, including offline evals, regression detection, and safety controls.

● Proficiency in multiple languages and frameworks (Python, Java, TypeScript, SQL).

● Strong cloud-native design experience on AWS (required), including serverless, containers, and managed data services.

● Hands-on experience with CI/CD and infrastructure-as-code (GitHub Actions, Docker, CloudFormation or similar).

● Solid data engineering fundamentals: data modeling, schema evolution, batch and streaming pipelines, and reconciliation strategy.

● Deep knowledge of observability practices (metrics, logs, tracing) and the judgment to instrument the right things.

● Ability to set technical direction across teams through influence, written design, and mentorship rather than authority.

● Strong written and verbal communication in English. You will work directly with internal stakeholders and vendor technical teams, and much of this role is specifying and documenting how systems and workflows behave.

● Judgment about which problems warrant an AI approach and which do not, supported by evidence.

Nice-to-Have Skills

● Familiarity with education interoperability standards (LTI 1.3 and LTI Advantage, OneRoster, Caliper, SCIM).

● Experience with durable workflow orchestration (Temporal, Step Functions, LangGraph, Prefect) and with MCP.

● Experience operating in regulated or air-gapped environments with restricted network egress.

● Background in education, justice, or mission-driven technology.

Benefits we offer

Access to e-Learning platforms.

Access to a virtual nutritionist

Amazing people-oriented organizational culture

Working from anywhere

Challenging projects using the latest technologies with clients from the US.