Staff AI Engineer

We're hiring an AI Engineer to build LLM-powered products that actually ship. You'll live at the seam between research and production - turning models, prompts, and retrieval into reliable systems users depend on. Less "train a model from scratch," more "make foundation models do useful work at scale, cheaply, and without hallucinating."

What you'll do

• Design and build LLM applications: RAG pipelines, agents, tool-use systems, structured generation, multi-step workflows
• Own prompt engineering and evaluation - write evals before you write prompts, and treat both as code
• Integrate foundation models (Claude, GPT, Gemini, open-weight Llama/Qwen/Mistral) and know when to reach for which
• Build the unglamorous infrastructure that makes LLMs production-grade: caching, streaming, retries, fallbacks, cost controls, observability, guardrails
• Fine-tune or distill models when prompting hits a wall (LoRA/QLoRA, SFT, DPO)
• Partner with product and design on UX patterns specific to AI (latency masking, streaming UIs, confidence display, human-in-the-loop)

What we're looking for

• 3+ years of software engineering, with at least 1 year shipping LLM-based features in production
• Strong Python and TypeScript; comfortable in a real codebase, not just notebooks
• Deep practical knowledge of the LLM stack: prompting techniques, function calling, structured outputs, context management, embedding models, vector search
• Have built and maintained an eval suite - and have opinions on why most public benchmarks are misleading
• Comfortable with at least one orchestration approach (custom, LangGraph, Inngest, Temporal) and one vector store (pgvector, Pinecone, Turbopuffer, Weaviate)
• Cost- and latency-aware: can read a token bill and a trace and know where to cut
• Clear communicator who can push back when "just add AI" isn't the right answer

Nice to have

• Experience with agent frameworks and MCP (Model Context Protocol)
• Fine-tuning experience on open-weight models, or work with inference servers like vLLM/TGI/SGLang
• Background in security, evals, or red-teaming for LLM systems
• Contributions to open-source AI tooling

What we offer

• Work from anywhere in the US! Machinify is digital-first.
  Top Medical/Dental/Vision offerings
• FSA/HSA
• Tuition reimbursement
• Competitive salary, 401(k) with company match
• Unlimited PTO
• Additional health and wellness benefits and perks
• Flexible and trusting environment where you'll feel empowered to do your best work

The salary for this position is based on an array of factors unique to each candidate: Such as years and depth of experience, set skills, certifications, etc. We are hiring for different levels and the base salary can range from $210k-$280k+ based on your assessed level. Compensation also includes meaningful equity, healthcare, unlimited PTO, and more.