Senior Software Engineer

Pearson’s AgentOps Engineering team is building enterprise agentic AI platform capabilities that teams across Pearson rely on. This Senior Software Engineer role is a hands-on opportunity to shape write-path capabilities, orchestration, routing, and long-running workflows that hold up in production.

What you’ll work on

You will help ship the platform features that let agents take real, governed actions. That includes wrapping enterprise APIs with validators and data sources so agents can write, not just read, while ensuring safe write semantics such as idempotency, confirmation and human-in-the-loop gates, structured outputs, and audit-ready action logging.

• Deliver write-path features for governed, enterprise actions through wrapped APIs, validators, and data sources.
• Design safe write behavior with idempotency, structured outputs, and audit-ready action logs.
• Build multi-agent and digital-worker orchestration patterns for delegation, collaboration, and multi-step completion.
• Create stateful, cyclic workflows that support reflection, recovery, and adaptive execution beyond linear chains.
• Develop the routing layer for intent classification, capability-based dispatch, fallback, and escalation paths.
• Build reusable components for retries, degraded modes, and human handoff, and tune agent roles and prompts using fixtures and golden sets.
• Work directly with the agentic platform via platform APIs to create, version, invoke, and debug crews, tasks, and graphs.
• Implement long-running, resumable workflows with checkpointing, persistence, and context restoration for non-deterministic AI.

How you’ll deliver

Delivery is primarily through AI-pair-programming, supporting high-velocity research, scaffolding, implementation, review, and shipping, while maintaining quality through TDD and rigorous verification. You will also help drive reusable engineering standards, shared libraries, and reference patterns, mentoring engineers through design and code reviews.

What you bring

• Hands-on senior experience building production-grade agentic systems.
• Independence to take complex workstreams from ambiguity to reliable, shipped software.
• Collaboration through standards and mentoring without relying on formal authority.
• Strong judgment across quality, latency, cost, resilience, and maintainability.
• Experience building LLM-powered systems, agents, or digital workers that run in production.
• Strong Python and backend/platform engineering including async services, typed code, and clean architecture.
• Fluency with agent orchestration frameworks, including building reliable agent tools with contracts, error handling, and tool chaining.
• Experience with routing/dispatch or workflow-control logic across components or services.
• Ability to use AI-pair-programming as a primary delivery mode without sacrificing quality.
• Prompt engineering expertise for structured outputs, nested schemas, and multi-agent coordination.
• Experience contributing to a real platform codebase (APIs, runtime, storage), not only an SDK.
• Solid APIs, distributed systems, and cloud-native engineering with production-reliability instincts.

Technology areas

• AgentOps Engineering, Python
• LLM-powered systems, AI-pair-programming, TDD
• Agent orchestration frameworks, async services, structured outputs
• Human-in-the-loop, enterprise APIs, tool chaining
• Checkpointing, persistence, context restoration

Benefits

• Access to cutting-edge tools, platforms, and thought leadership.
• A collaborative, inclusive, and innovation-driven culture.
• Real ownership of orchestration, routing, and platform patterns others adopt.
• A front-row seat building enterprise agentic AI infrastructure that teams across Pearson rely on.
• Competitive benefits designed to support the diverse needs of people and their families.
• This position is eligible to participate in an annual incentive program.

Even better if you have

• Experience evaluating agents using task success, groundedness, tool-use accuracy, schema conformance, and regression against golden fixtures.
• Experience with tool/context interoperability protocols such as MCP.
• Experience with a major cloud platform (AWS preferred), containerization, and CI/CD, plus familiarity with state stores for orchestration and persistence.
• AI observability and evaluation tooling for LLM systems.
• RAG and memory patterns including vector databases, hybrid retrieval, re-ranking, and grounding.
• Secure execution, sandboxing, and prompt-injection mitigation.

Location: Remote (remote). Compensation: USD 90,000 - 135,000 per year.

Application window: Applications will be accepted through 5th August 2026. This window may be extended depending on business needs.

Why Pearson: At Pearson, we accelerate careers and support internal mobility and leadership development. Compensation is influenced by factors including skill set, experience, and location.