Shepherd
Founding Machine Learning Engineer
$180K–$400K
San Francisco · On-site · Full-time · Posted Apr 2, 2026
- Experience
- 4+ years of industry experience.
About the job
Shepherd’s first Machine Learning Engineer builds and operates end-to-end ML systems that advance autonomous underwriting, from raw data and model development through production, evaluation, and monitoring.
What you'll do
- Build and ship production ML systems and feedback loops that improve autonomous underwriting decisions.
- Develop confidence scoring, evaluation, observability, monitoring, and guardrails for safe AI underwriting.
- Work with underwriters and cross-functional teams to translate domain expertise into reliable, auditable LLM workflows.
What they ask for
- 4+ years of industry experience building and shipping ML systems end-to-end, including experience with model deployment platforms.
- Experience fine-tuning SLMs/LLMs and using LLMs in production, including prompt engineering, structured outputs, tool use, and evaluation.
- Deep proficiency in Python and modern ML frameworks.
- Experience building reliable models with limited labeled data, including synthetic data generation or data augmentation.
Nice to have
- Familiarity with document parsing, information extraction, or NLP on unstructured business documents.
- Experience in insurance, finance, or other high-stakes structured domains.
Pay
USD 180K – USD 400K base salary
Tools
Python, AWS Sagemaker, PyTorch, HuggingFace, Tensorflow, OpenAI Gym/Gymnasium, LangChain, LangGraph
AI focus
Build ML systems to advance underwriting autonomy from L1 to L3 and beyond, including agentic LLM workflows, human-feedback training loops, confidence scoring, evaluation, and safety infrastructure.
Model-written summary · the original posting is authoritative