I

Innovaccer

Healthcare data analytics platform.

Principal AI Researcher

AI ResearchConfirmed live in the last 24 hours
Full-timeSan Francisco, CA, USANo salary listedIn-person

About the Job

Requirements

  • A master's or doctoral degree in Computer Science, Machine Learning, or a related quantitative field; exceptional bachelor's candidates with substantial research or open-source work will be considered.
  • Depth beyond coursework demonstrated through first-author publications at NeurIPS, ICML, ICLR, ACL, EMNLP, or similar venues; meaningful open-source machine-learning contributions; an AI-lab research internship; or models personally trained and shipped for real users.
  • Personal experience fine-tuning an open-weight model, understanding parameter-efficient versus full fine-tuning, and explaining the choice made.
  • Strong Python and PyTorch skills, and familiarity with training and serving tools such as Hugging Face, FSDP or DeepSpeed, and vLLM or SGLang, or equivalents.
  • Some exposure to multi-GPU training, including understanding sharding strategies and their importance.
  • Evidence of completing work.
  • Experience post-training models that ran in production and understanding what failed.
  • Experience building the model layer behind a real product feature, including fine-tuning at multiple model sizes and composing models into a system meeting production accuracy and latency requirements.
  • Hands-on multi-node training experience using at least tens of GPUs on models with tens of billions of parameters or more.
  • Experience building or substantially owning a data pipeline for a real training run, including deduplication, filtering, decontamination, and format normalization.
  • Ability to scope an ambiguous problem into a plan and distinguish research questions from engineering tasks.
  • Deep familiarity with large-scale distributed training failure modes and the operational discipline needed to keep a long training run active.
  • A public or verifiable record of papers, model releases, systems, or a training program with demonstrated results.

Responsibilities

  • Work with product, engineering, and data teams to design, develop, and deploy AI-powered applications, including large language model-based solutions, AI agents, retrieval-augmented generation, and intelligent automation workflows.
  • Take ideas from research papers through prototype and production.
  • Build the model layer of a product by selecting model sizes, combining models into a working system, and meeting product-level accuracy requirements.
  • Write code in Python and PyTorch and identify causes of slow training runs.
  • Design experiments by stating hypotheses, running ablation studies, and reporting results, including results that challenge the hypothesis.
  • Build evaluations before models and measure results critically.
  • Read current research and assess which techniques are likely to be robust.
  • Explain work to non-AI engineers, including clinicians and operators, and incorporate their feedback.
  • Set technical direction over a two-to-four-quarter horizon and enable a team to execute it.

Desired Qualifications

  • Mentoring other engineers or setting direction for them is a plus and will become more important as the team grows.
  • Experience leading post-training on models with hundreds of billions of parameters across runs using hundreds of GPUs, with ownership of the outcome.
  • Experience designing the complete model architecture for a shipped product across small, medium, and large models, including economic decisions about model tiers and composition.
  • Experience making model-program decisions about data mixture, objective design, when to stop a run, and when to abandon an approach.

About the Company

Innovaccer

Healthcare data analytics platform.

Innovaccer provides a Health Cloud platform that connects, curates, and analyzes healthcare data to improve patient outcomes and reduce costs. The platform lets healthcare providers manage transitional care, lower 30-day readmissions, and expand primary care services, while giving pharmaceutical companies like Roche contextualized patient insights to lower development costs and risk. Revenue comes from clients paying for platform use. Unlike others, Innovaccer demonstrates value through measurable results (e.g., a CHESS $3 million value and a 23% reduction in 30-day readmissions) and earns high scores in independent analytics rankings (e.g., 94.9/100 in KLAS Data & Analytics Platforms with “A” across all six Customer Experience Pillars). The company's goal is to turn healthcare data into practical, scalable insights that improve clinical and financial outcomes for providers, life sciences partners, and patients.

Company size
1,001-5,000
Company stage
Private
Total funding
$653.1M
Headquarters
San Francisco, CA
Founded
2012

Recently Posted Jobs

Explore more opportunities at Innovaccer

See more jobs