Principal AI Engineer
Remote
Job Summary
Own architecture for complex Generative AI and agentic systems end-to-end, setting standards for evaluation, observability, and responsible AI. Stay in the code for foundation-model fine-tuning, novel agentic workflows, and advanced RAG pipelines using Python on Google Cloud. Engineer for production by designing for latency, reliability, cost, and scale, applying MLOps discipline to ensure systems reach production. Architect multi-step reasoning systems that automate complex reasoning reliably and advise clients directly to translate business strategy into AI architecture. Elevate senior and mid-level engineers through architecture reviews and mentorship to establish a high bar for AI-augmented engineering.
Required Qualifications
- Bachelor's or Master's degree in Computer Science, Engineering, or a related technical field
- 10+ years in software / AI / ML engineering, with a substantial track record of AI systems delivered to production at scale
- Demonstrated technical leadership — owning architecture and setting direction across engagements or teams, not just individual deliverables
- Proven track record of deploying GenAI and/or agentic products to production environments
- Strong data engineering and SQL knowledge
- Senior client-facing experience — translating technical complexity into business value for executive stakeholders
- Mastery of Python and shell scripting
- Fluency across the modern AI engineering stack
- Deep, current expertise with Google Gemini, GPT-class, and open models (LLaMA)
- Advanced prompt engineering, fine-tuning, and evaluation
- Proven experience designing and productionizing agentic and multi-step reasoning systems (MCP, tool use, orchestration)
- Expertise in vector databases (Vertex AI Vector Search, pgvector, Pinecone) and semantic search at production scale
- Deep hands-on experience with Google Cloud / Vertex AI and architecting scalable, resilient software systems
- Strong command of LangChain, LlamaIndex, or equivalent orchestration layers
- A first-class software engineer — clean, maintainable code, full SDLC ownership, and the architectural judgment to lead others
Desired Qualifications
- Experience with classic machine learning (neural nets, training, tuning)
- foundation-model or novel-model work
Hiring someone like this?
Get your role in front of qualified candidates on Sorce.