Senior AI Engineer

Your Role

We're seeking a Senior AI Engineer to join our team. You will build the AI products that sit directly between a job seeker and their next, better job. This includes:

• LLM-based systems like chat-based coaching that meets users where they are,
• Agentic workflows that complete multi-step tasks (e.g., completing eligibility forms, assembling applications), and
• Orchestration systems that pull accurate context from case management systems and other data sources.

The centerpiece of this role is our coaching products. You will guide job seekers to their next best step, grounded in evidence from labor market data and interventions that have been proven to lift wages and lifetime earnings. You will own both the evidence-based analysis and the engineering that turns it into a product.

This is a hands-on, high-ownership role on a small team. You will partner closely with Engineering, Product, and the VP of Data & AI, with wide latitude to decide how these systems get built.

What You'll Own

• LLM product development: Design, build, and ship end-to-end LLM-based product experiences, including conversational coaching, agentic workflows, and orchestration layers.

• Coaching and pathways: Build products grounded in evidence, including observed career transitions, causal impact on wages, and honest assessments of data limitations.

• Evaluation and quality: Own how we measure AI effectiveness, including eval harnesses, offline/online quality measurement, regression tracking, and human-in-the-loop reviews.

• Applied ML beyond LLMs: Build predictive, classification, and ranking models when appropriate, and know when to use an LLM.

• Data engineering for AI: Prepare, label, govern, and maintain pipelines and knowledge sources for AI systems, ensuring retrieval and reasoning are grounded in trusted data.

• Integration engineering: Connect AI systems safely to internal tools, databases, SaaS products, case management systems, and enterprise workflows with proper auth, guardrails, and failure handling.

Where This Role Can Go

This role carries real influence over how we build AI systems. You will shape patterns in prompting, evaluation, and agent behavior while building brand-new products. Future paths include deeper technical leadership over the AI platform or broader ownership of coaching and pathways products.

Required Experience

• 5+ years of applied ML/AI engineering experience, with a track record of shipping systems into real products used by real people.

• Demonstrated LLM product experience, including prompt engineering, agentic workflows, tool use, retrieval, and orchestration in production (not just prototypes).

• Real evaluation discipline: ability to measure LLM system performance, build evals, define quality metrics, catch regressions, and speak concretely about system failures.

• Classical ML depth: solid grounding in machine learning fundamentals and hands-on experience building and rigorously evaluating predictive or classification models.

• Fluency in Python and SQL, with real comfort in production codebases.

• Experience building data and integration foundations for AI, including ingestion pipelines, labeling, knowledge sources, governance, and connecting to external systems.

• Sound instincts about safety and guardrails in consequential domains, where confidently wrong answers have real costs.

• Clear communication skills to explain models, agent behavior, and evaluation results to non-technical audiences.

Bonus Points

• Experience with AWS and SageMaker, or deploying AI systems on AWS.

• Experience with jobs-and-skills, HR, or labor market data, or skills/occupation frameworks like O*NET or ESCO.

• Experience integrating with case management systems, government systems, or public-sector data environments.

• Experience with recommender systems, ranking, or search at scale.

• Experience with retrieval-augmented generation at production scale, including retrieval quality evaluation.

• Experience with labor economics, program evaluation, or measuring the impact of workforce interventions.

• Publications, talks, blog posts, or open-source work showing depth in AI or ML engineering.

Our Tech Stack for Data & AI

• Languages: Python, SQL

• AI and LLM: Modern LLM and agent tooling, including agent orchestration frameworks (e.g., LangChain or LangGraph), vector search, and RAG tooling.

• Machine learning and NLP: scikit-learn, modern NLP and embedding tooling, AWS SageMaker

• Orchestration and transformation: Airflow, dbt

• Storage and warehousing: PostgreSQL, Redshift, MongoDB

• Cloud: AWS

• Visualization and reporting: Looker, Quicksight