Sr Software Engineer - AI

McGraw Hill is creating intelligent learning experiences through its AI Platform team, focusing on building AI that enhances education. The Senior Software Engineer - AI will design, develop, and deliver AI-powered products, collaborating with various teams to turn innovative research into practical applications.

Responsibilities:

• Build and ship AI-powered products: Design, develop, and maintain generative AI applications — from RAG pipelines with vector search and semantic chunking to LLM orchestration, production APIs, and async task workflows. You'll own features from concept through deployment and monitoring, making pragmatic architectural decisions that balance innovation with reliability at scale
• Operate a dual-language backend: Work across Python (FastAPI) and Go (Gin) microservices, understanding when each language's strengths apply
• Build async-first APIs, background workers (Celery/SQS), and data pipelines that handle millions of concurrent users
• Own infrastructure and observability: Contribute to Terraform modules, Kubernetes manifests (Kustomize), and CI/CD pipelines (GitHub Actions). Instrument services with our observability stack (New Relic, Datadog, Prometheus) so the team can ship with confidence and debug production issues quickly
• Lead through technical excellence: Drive design reviews, write clear technical documentation, and make thoughtful decisions about tradeoffs. Improve team standards around code quality, testing, observability, and system reliability. You're the engineer others come to when they're stuck on a hard problem
• Mentor and multiply: Support the growth of engineers on the team through pairing, code review, and design guidance. Share knowledge proactively and foster an inclusive, learning-oriented team culture
• Collaborate cross-functionally: Partner with data scientists to evaluate model performance and fine-tune prompts. Work with product managers and designers to translate product vision into technically sound solutions. Communicate tradeoffs clearly and build trust across teams
• Stay at the frontier: The AI landscape moves fast. You'll experiment with new models, frameworks, and techniques — and bring a point of view on which advancements are worth integrating into our platform and which are noise
• Ship with confidence: Build with accessibility in mind, meeting WCAG 2.2 AA standards. Champion code quality through static analysis, type checking, security linting, and comprehensive test suites

Requirements:

• 5+ years of professional software development experience, with a strong track record of delivering and operating distributed systems
• Practical experience building generative AI applications — working with LLMs (Azure OpenAI or similar), prompt engineering, RAG architectures (vector databases like Pinecone, embedding models, semantic chunking and routing), and custom orchestration pipelines
• Deep proficiency in Python (FastAPI, async/await, Pydantic) and working knowledge of Go (or willingness to become proficient quickly)
• Experience building and maintaining microservices, async task workers (Celery/SQS), and APIs that power AI capabilities
• Comfortable with PostgreSQL (including Aurora and read/write splitting patterns)
• Production experience with AWS (ECR, RDS Aurora, S3, SQS, IAM) and Azure (Azure OpenAI, private endpoints)
• Hands-on with Terraform for infrastructure-as-code and Kubernetes (Kustomize, HPA) for container orchestration
• Experience with APM and monitoring tools (New Relic, Datadog, or similar)
• You value code quality — static analysis, type checking, security linting, and comprehensive test suites (unit, integration, performance)
• You identify problems, propose solutions, and drive them forward
• Strong communication and collaboration skills
• You're excited about the intersection of AI and education — and motivated by the idea that the systems you build will directly help students learn
• Familiarity with semantic routing and semantic chunking libraries for RAG pipelines
• Experience with k6 or similar performance/load testing frameworks
• Background in edtech, adaptive learning, or content-rich product domains
• Experience with evaluation frameworks for LLM outputs (automated scoring, human-in-the-loop review)
• Familiarity with agentic AI patterns and multi-step reasoning workflows
• Experience implementing accessible features in web applications (WCAG 2.2 AA)
• Familiarity with secrets management patterns (SOPS, AWS KMS)