Principal AI Engineer

Location: Remote (US). Texas preferred for occasional in-person collaboration.

Compensation: $160,000 to $180,000 base, plus equity

About the Role

We're hiring a Principal AI Engineer who lives and breathes applied AI. You've shipped real LLM-powered systems, not just experimented with them. You think in pipelines, obsess over eval design, and know firsthand how much the details matter when putting agents into production.

This role sits at the intersection of engineering and applied research. You'll lead the design and implementation of core AI infrastructure, including RAG pipelines, evaluation frameworks, and multi-agent systems, and help define how the team builds with LLMs as the technology evolves.

The product processes complex documents in regulated industries where accuracy and reliability are paramount. Your work will directly shape how legal and medical professionals turn dense, messy source material into actionable intelligence.

Track record matters more than pedigree. Side projects, open-source contributions, and personal systems count just as much as anything on a résumé.

What You'll Do

1. Design, build, and own production-grade RAG pipelines, including chunking, embedding strategies, retrieval optimization, and re-ranking
2. Architect and implement LLM evaluation frameworks: automated evals, human-in-the-loop review, and regression testing across model versions
3. Build and maintain multi-agent systems using modern agent frameworks with a focus on reliability, observability, and cost efficiency
4. Own LLM observability using LangFuse, including tracing, cost monitoring, latency analysis, and quality tracking
5. Define best practices for prompt engineering, context management, and tool use across the team
6. Evaluate and integrate new models, frameworks, and tooling as the ecosystem evolves
7. Collaborate across the full stack. The AI layer is deeply integrated with a React/NestJS product and an async microservices pipeline
8. Mentor other engineers and establish patterns that scale

What We're Looking For

1. 8+ years of software engineering experience with a meaningful portion focused on applied AI/ML systems
2. A GitHub and side projects outside of work. We care about what you build, not just where you've worked
3. LLM application development. Real products on top of GPT, Claude, Gemini, or open-source equivalents
4. RAG systems. Retrieval pipelines, vector databases (LanceDB, Pinecone, pgvector), hybrid search, re-ranking, query rewriting
5. Evals. Designing and running evaluations for LLM outputs, custom eval harnesses, quality tracking
6. LLM observability. LangFuse, LangSmith, or Braintrust for tracing, monitoring, and debugging at scale
7. Agents and agent frameworks. LangGraph, CrewAI, AutoGen, or similar, including production deployments
8. Multi-model architectures. Routing across models and orchestrating specialized agents
9. Async or event-driven systems. AI workflows on message brokers (Azure Service Bus a plus)

Stack

1. Platform: Azure-based microservices
2. Product: React, TypeScript, NestJS
3. AI: LanceDB for retrieval, pub/sub messaging for async workflows, LangFuse for observability

Nice to Have

1. Microservice or distributed pipeline architectures (event-driven or DAG-based)
2. Open-source contributions or public projects in AI/ML
3. Fine-tuning or model adaptation (LoRA, RLHF, DPO)
4. Azure AI services (OpenAI, Document Intelligence, AI Search)
5. Structured outputs, function calling, and tool use

If you're the kind of engineer who pushes back when someone says "just make it agentic," who designs for failure modes before they happen, and who can explain to a non-technical stakeholder why a prototype isn't production-ready, this is your role.

Are you looking for remote jobs near your area? At Yulys, thousands of employers are looking for exceptional talent like yours. Find a perfect job now.