Senior AI Product Engineer

Just Right Reader•Remote,

2d•$180 - $200

About The Position

Just Right Reader is a curriculum provider focused on advancing literacy for students from Birth through 12th Grade. Its resources are grounded in the Science of Reading and support learners, including older students who need intervention to build foundational skills. The company works with school districts and state education agencies to provide students with books at the appropriate reading level. Its platform uses TypeScript, Next.js, and PostgreSQL and connects with CRM, ERP, and internal business systems. The Senior AI Product Engineer will build and ship planned AI product projects that have not yet been started. The role involves making technical decisions about model calls, handling incorrect outputs, managing cost and latency, and testing AI features. Establishing evaluation methods early is important so the team can detect whether prompt or model changes improve or degrade results. Since the product serves K–12 students, privacy and data handling must be considered throughout design, and recommendations need to be defensible. The engineer will also contribute to regular product development, including Next.js work, schema design, and integrations. This is a small-team role involving direct conversations with operations, customer success, and fulfillment colleagues, as well as internal tooling. The company is not seeking model training, fine-tuning, or research work.

Requirements

Experience building an AI product from scratch and shipping it to real users, including making architecture decisions.
Senior-level experience as a full-stack or product engineer on production systems with real users and meaningful consequences when systems fail.
Strong TypeScript skills across application and service development.
Production experience with Next.js, including the App Router, server components, and API routes.
Strong backend skills, including API design, server-side business logic, and cross-system debugging.
Strong PostgreSQL skills, including query design and optimization, schema and data modeling, and knowing when to use SQL directly rather than an ORM.
Experience integrating third-party APIs and handling failure cases such as retries, idempotency, rate limits, and partial failures.
Application-level AWS experience with services such as S3, SQS, Secrets Manager, ECS/Fargate, Lambda, RDS, or DynamoDB, including debugging running systems.
Understanding of production LLM failure modes, including malformed structured output, schema drift, latency spikes, cost surprises, refusals, and confidently incorrect answers.
Experience evaluating AI quality, such as with golden sets, scored runs, human review, or pre-deployment regression checks.
Judgment about when a rules-based or SQL-based solution is more appropriate than a model.
Awareness of AI operating costs and ways to manage them.
Ability to communicate effectively with colleagues in operations, customer success, and fulfillment.

Nice To Haves

Production experience with retrieval systems.
Experience building or seriously using evaluation tooling.
Prisma experience.
Terraform or general infrastructure literacy.
Python experience.
Previous K–12 or FERPA experience.
Experience with NetSuite, HubSpot, Shopify, or Slack.

Responsibilities

Build and ship the company's planned AI product projects.
Make technical decisions about model invocation, handling incorrect or malformed outputs, cost, latency, and testing.
Establish evaluation and regression-checking practices to assess whether prompt or model changes improve or worsen results.
Design AI features with privacy, data handling, and defensible recommendations in mind, given the K–12 context.
Contribute to regular product engineering, including Next.js development, schema design, and integrations.
Build and maintain production systems involving APIs, backend business logic, PostgreSQL, and third-party integrations.
Work directly with operations, customer success, and fulfillment users to understand and support their needs.
Treat internal tooling as a normal part of product engineering work.