Staff Machine Learning Engineer

Mural Group; United States; Remote; USD 230,000 - 300,000; Full time

Job summary

Mural Group is hiring a Staff ML Engineer to set the technical direction for its agent systems, working on model and systems that ship to large enterprise customers rather than wrapping someone else’s API.

You will design multi-step LLM systems, own evaluation from offline benchmarks to online experiments, and fix failure modes like hallucination and tool misuse, while mentoring senior engineers and shaping team practices.

Qualifications

• 7+ years shipping ML in production, including LLMs at scale.
• Led the design of complex, multi-step LLM systems.
• Depth in transformers, embeddings and retrieval.
• Evaluation experience beyond single prompts.
• BS in CS or a related field; MS or PhD a plus.

Responsibilities

• Design multi-step LLM systems: reasoning, retrieval, planning and tool use.
• Own evaluation, from offline benchmarks to online experiments and model-graded review.
• Find and fix failure modes such as hallucination and tool misuse.
• Set the quality, latency and cost targets the team builds against.
• Mentor senior engineers and shape how the group works.

Skills

LLM systems design

Production ML

Evaluation pipelines

Distributed cloud

Mentoring

Education

BS in CS or related field

MS or PhD a plus

Tools

PyTorch

Job description

Location: Remote (US)

Pay: $230,000-$300,000 + bonus + equity

A late-stage, venture-backed enterprise AI company is hiring a Staff ML Engineer to set the technical direction for its agent systems. This is model and systems work that ships to large enterprise customers, not a thin layer over someone else's API.

What you'll do

• Design multi-step LLM systems: reasoning, retrieval, planning and tool use

• Own evaluation, from offline benchmarks to online experiments and model-graded review

• Find and fix failure modes such as hallucination and tool misuse

• Set the quality, latency and cost targets the team builds against

• Mentor senior engineers and shape how the group works

What you bring

• 7+ years shipping ML in production, including LLMs at scale

• Led the design of complex, multi-step LLM systems

• Depth in transformers, embeddings and retrieval

• Evaluation experience beyond single prompts

• PyTorch or similar, on distributed cloud infrastructure

• BS in CS or a related field; MS or PhD a plus

Nice to have:

ML shipped before LLMs, then moved into generative AI; detailed retrieval work (chunking, reranking, evals)