Uses AI-native tools and agent frameworks for fast prototyping and delivery; emphasizes AI coding tools and eval/observability for trustworthy AI.
About the Role
As an Applied AI Engineer on JD Power's Launch Lab team, you build and ship full-stack AI-powered applications from PoC blueprints to production-quality interfaces on 2–4 week sprints. You own architecture, deployment, observability, and clean handoffs while ensuring AI behavior is reliable, cost-aware, and integrated with core data infrastructure.
Job Description
Role
Applied AI Engineer (P4) on the Launch Lab team delivering full‑stack AI-powered applications (internal tools, lightweight client-facing prototypes, integration accelerators) on rapid 2–4 week sprint cycles. Own PoC blueprints end-to-end: implement production-quality interfaces and services, deploy and operate them, ensure observability and cost controls, and hand off to maintaining teams.
Key Responsibilities
- Lead application architecture and implementation across Launch Lab delivery tracks; deliver frontend and backend for polished, production-quality interfaces.
- Translate PoC Blueprint Handoffs (Power Agents, vector DB hooks, agent workflows) into production applications with measurable outcomes and evaluation harnesses.
- Consume internal agent infrastructure (Power Agents, Agent Skills Registry) and surface Intelligence layer capabilities through user-facing apps.
- Integrate outputs with core data infrastructure (Snowflake, internal APIs, API Gateway) following gateway standards and embedding AppSec, Data Governance, and PII handling in designs.
- Participate in Gate 2 technical scoping: estimate effort, identify dependencies, define success criteria and cost budgets before building.
- Own deployment, confirm operational availability, and execute clean handoffs with usable documentation for maintaining teams.
- Contribute to Emerging Technology Radar and Architecture Decision Records; run lightweight user testing and postmortems.
Requirements
- Portfolio of shipped, production AI applications with real-user impact and experience diagnosing/fixing production failures.
- Full-stack engineering depth: frontend proficiency (React preferred) and backend experience across REST and/or GraphQL, authentication patterns, and cloud-native service architecture.
- Hands-on experience with agentic frameworks or LLM APIs (examples listed: LangChain, LlamaIndex, Anthropic, OpenAI) and RAG patterns; experience with streaming responses and building intuitive AI interfaces.
Also Valued
- Eval harness design and LLM observability experience (Langfuse, LangSmith, Braintrust or equivalent).
- Multi-model routing and inference cost optimization experience (LiteLLM, Portkey or equivalent).
- Data platform experience (Snowflake, BigQuery) and security basics for AI applications (PII handling, secrets management, prompt injection defense, API audit logging).
- Strong product intuition, documentation and postmortem practice, and familiarity with JD Power internal platforms and data ecosystem.
Logistics
- Location: Remote – USA, Canada, Europe or Australia. Virtual-first company with flexible work arrangements.
- Compensation: Starting salary range listed (see posting). Candidates must be legally authorized to work where employment is offered; no employment sponsorship provided for this position.