Pearson Virtual Schools serves K-12 students and their teachers on Connexus, a large-scale student, curriculum, and assessment data platform. You will manage engineers building and operating that platform and the products on top of it.

We are modernizing a platform that today runs primarily on legacy .NET Frameworks: IIS-based ASP.NET WebForms and MVC, an AngularJS and jQuery front end, and SQL Server. The future is design-system-based React and TypeScript single-page applications calling .NET REST services that operate in containerized modern .NET, and integrate with an API-first, event-driven model on AWS with externalized authorization. Your teams will work on both sides of this technology, the legacy and the new.

The platform is becoming agentic across every exposed product area, not only in the teams whose names say AI. That changes what an engineering manager is accountable for. Wherever your team's surface reaches a student, a family, a teacher, or school staff, you own the guardrails, the human-in-the-loop fallback, the disclosure, and the evidence that the behavior was evaluated before release. That obligation does not vary by team.

You are expected to be AI-native in how you manage. That means you can read the work of engineers operating at our bar for AI-fluent engineering, you know what an evaluation set is worth and what it is not, and you hold quality as a gate rather than a phase. A good deal of the code your team is accountable for will not be typed by hand.

Key Responsibilities

Team Delivery and Operating Rhythm

Allocate daily work, provide immediate support, and keep the delivery schedule reliable

Give feedback in the moment, and escalate an unresolved issue early rather than absorbing it

Own onboarding and ramp for new team members, and shorten it deliberately

Holding the Shape of the Team

Hire against our bar for AI-fluent engineering, and hold it when the pipeline is thin

Agentic Capability and Duty of Care

Treat every exposed product surface as one that will carry agentic capability, and apply the same obligations whether or not the team's name says AI

Require an evaluation set with a stated pass bar, wired into the release gate, before model-backed behavior reaches a student

Ensure guardrails, human-in-the-loop fallback, escalation, and disclosure are designed into the interaction rather than added after a complaint

Monitor reliability, hallucination, and model performance against stated thresholds, and treat drift as a defect rather than as news

Quality, Adoption, and Shared Tooling

Hold specification-first practice: contract-level tests before the implementation, negative tests as a first-class obligation, and evals wherever behavior is model-backed

Ensure the automated merge gates run and are not routinely overridden, and treat a suite that never fails as a defect rather than a result

Drive the team's contribution to the governed shared library, so skills, MCP servers, evaluation suites, and golden datasets leave the team rather than staying in it

People and Capability

Coach on the job, in the work, rather than in a scheduled session about the work

Hold process adherence where it matters and challenge it where it does not

Recognize contribution specifically, so the team knows what good looked like

Domain Depth for This Posting

Build and maintain team-level understanding of Connexus data foundations, including student, enrollment, curriculum, assessment, progress, attendance, and school operations data, so delivery decisions are grounded in the domain.

Lead the team's migration and cutover execution for assigned domains, including legacy behavior analysis, reconciliation, rollback planning, dependency coordination, and production-readiness validation.

Guide engineers in translating legacy SQL Server, stored procedure, and .NET behavior into durable PostgreSQL schemas, APIs, events, and domain contracts, while ensuring designs are reviewed and understood across dependent teams.

Apply policy-aware data access expectations for K-12 education systems, including FERPA-sensitive data boundaries, externalized authorization, and audit-ready controls, through planning, delivery reviews, and release readiness.

Preferred Qualifications

2+ years managing engineers who ship to production, including hiring, performance, and development

A working engineering background, enough to review a technical decision and to be credible with engineers operating at our bar for AI-fluent engineering

Experience delivering a model-backed or agentic capability to real users, including how it was evaluated before release and monitored afterward

Practiced at holding a quality gate under delivery pressure, and able to describe a time you held one and a time you should have

Experience managing across a legacy and a target platform at the same time, where the team cannot stop shipping on either

Comfortable managing a distributed team across time zones, with the written discipline that requires

Experience in regulated or privacy-sensitive environments; K-12 and FERPA exposure an advantage

Relevant degree or equivalent practical experience

Skills and Experience

The skills and experience this role is assessed against.

Core skills

Adaptability

Artificial Intelligence

Learning to Learn

Engineering fundamentals

Collaboration

Information Security

Innovation

Problem Solving

Technical Acumen

Role-specific skills

API Management

Cloud Computing

Data Security

DevOps

Ethical Responsibility

Resilience Planning

Software Design

Emerging skills

AI Agents

Automated Artificial Intelligence (AutoAI)

Specialist depth for this role

Large Language Models (LLMs)

Generative AI

Evaluation of AI-Generated Content Quality

Performance Management

Code Review

Overseeing delivery of AI-enabled capabilities and holding the release gate

Measuring and reporting team adoption of AI tooling

Building reusable engineering tooling and evaluation suites that other teams adopt