Staff Machine Learning Engineer - Llms & Document AI

EvenUp EvenUp · Vertical AI · San Francisco, CA · Hybrid · Data Science

Staff Machine Learning Engineer focused on building and deploying advanced Document AI models and LLM fine-tuning for a legal tech SaaS company. The role involves solving complex modeling challenges, establishing evaluation standards, and providing technical leadership.

What you'd actually do

  1. Design and refine advanced Document AI models for entity/relationship extraction, document structure understanding, and sophisticated reasoning from complex legal and medical text.
  2. Solve complex modeling challenges involving long-context and multi-document reasoning, including context segmentation and the aggregation of distributed facts.
  3. Lead LLM fine-tuning initiatives, applying techniques like reinforcement learning with verifiable reward signals and parameter-efficient fine-tuning (e.g., LoRA, QLoRA).
  4. Establish rigorous evaluation standards to reduce hallucinations, improve factual consistency, and handle ambiguous or noisy data.
  5. Drive data excellence by conducting hands-on analysis to ensure high-quality training and evaluation datasets, managing edge cases, noise, and data drift.

Skills

Required

  • Python
  • LLM technologies
  • ML/NLP frameworks
  • Deep learning
  • Reinforcement learning
  • Probabilistic modeling
  • Optimization
  • Entity/relationship extraction
  • Document structure understanding
  • Long-context reasoning
  • Multi-document reasoning
  • LLM fine-tuning
  • Parameter-efficient fine-tuning (LoRA, QLoRA)
  • Prompt engineering
  • Data analysis
  • Technical leadership
  • Mentorship

Nice to have

  • PhD in Machine Learning
  • Computer Science
  • Quantitative fields
  • EvenUp's mission
  • Hybrid work

What the JD emphasized

  • core to our customer experience and business growth
  • high-priority role
  • shape our modeling strategy
  • owning critical areas
  • drive innovation
  • set the vision
  • tangible product launches
  • robust ML foundation
  • direct influence
  • build the future
  • lead the development
  • complex legal document challenges
  • production-ready models
  • advanced Document AI models
  • sophisticated reasoning
  • long-context and multi-document reasoning
  • LLM fine-tuning initiatives
  • reinforcement learning
  • parameter-efficient fine-tuning
  • rigorous evaluation standards
  • reduce hallucinations
  • improve factual consistency
  • data excellence
  • high-quality training and evaluation datasets
  • advanced prompt engineering techniques
  • technical leadership and mentorship
  • fostering a culture of technical excellence and continuous growth
  • translate ambiguous research goals into impactful production solutions
  • bridge between cutting-edge research and practical application
  • A true builder’s mentality
  • launch, scale, and shape a new technical domain
  • Deep domain expertise
  • track record of solving complex modeling challenges and deploying models in operational settings
  • Expertise in advanced ML techniques
  • Strong record of mentorship
  • guiding team members
  • establishing best practices for experimentation and benchmarking
  • high proficiency in Python
  • major LLM technologies
  • Excellent communication skills
  • deliver business-impactful solutions
  • 5+ years of hands-on professional experience
  • multiple models deployed in operational settings
  • Strong proficiency with the latest LLM technologies
  • demonstrated ability to apply cutting-edge research to practical solutions
  • High proficiency in Python
  • deep understanding of modern ML/NLP frameworks
  • Demonstrated ability to lead technical strategy
  • mentor team members
  • drive execution
  • fast-paced, ambiguous environments
  • Experience working in a high-growth startup environment
  • Excellent cross-functional leadership skills

Other signals

  • LLM fine-tuning
  • Document AI
  • information retrieval
  • long-context reasoning
  • reduce hallucinations
  • factual consistency