Job description

BUILDING HEALTHCARE THAT REMEMBERS Every time you call a doctor's office, the system forgets you the moment you hang up. You wait on hold, explain everything from scratch, get bounced between lines, and hope someone follows up. Everyone has felt this. No one should have to. At Assort Health, we're building one continuous conversation for every patient. An AI agent that knows who you are, remembers how you like to be reached, and carries that context from your first symptom to your full recovery. What began as the first voice AI agent to schedule a specialty appointment has become the largest proprietary specialty dataset in healthcare: 285 million patient interactions, 62,000 care protocols, 1.6 million decision pathways. We went from 15 people in 2025 and will cross 250 by the end of 2026. Revenue is up 20x in 15 months and we've raised $222 million, including a recent $120 million Series C assorthealth.com at a $1.2 billion valuation. We've also been named to the Forbes Cloud 100 and the 2026 Enterprise Tech 30 as the only healthcare AI company on the list! Healthcare spends $1.1 trillion a year on administration instead of care. If you want to help change that, apply below. ABOUT THE ROLE As a Research Engineer, you’ll be responsible for building industry-leading healthcare AI models and model pipelines that power Assort’s voice agents and overall patient platform, taking them all the way from idea to production. You’ll own large initiatives that push the boundary of what models and agents can accomplish in the healthcare space while maintaining high levels of reliability, efficiency, and precision of accomplishing tasks. We’re looking for strong research engineers who want to build the core models and algorithms behind our core platform and agents. People here own their work end-to-end, ship real improvements, and are trusted to make high-impact technical decisions. WHAT YOU’LL DO - Lead research and engineering efforts to improve core conversational capabilities in production, including instruction following, tool calling, retrieval, and memory - Investigate research questions in speech and audio, and design experiments to test hypotheses about model behavior and capabilities - Build and iterate on end-to-end models and pipelines that optimize for quality, efficiency, and user experience - Identify opportunities for model-task optimization to ensure the right model architectures are used for different tasks based on performance, latency, and cost - Partner with platform and product engineers to integrate new models into production systems - Break down ambiguous research ideas into clear, iterative milestones and roadmaps - Train, fine-tune, validate, and further develop models built in-house at Hume and models developed by our customers - Design evaluation protocols and benchmarks that measure model quality, generalization, robustness, and progress against research objectives - Analyze datasets, model outputs, and failure cases; compare training approaches and use the evidence to guide the next round of model development WHAT WE’RE LOOKING FOR - 5+ years of experience with a minimum of 3+ in AI/ML engineering or research - Prior experience post-training and deploying LLMs in production environments - Fluency in Python and modern ML tooling (training, evaluation, inference, data pipelines) - Track record of taking research ideas from prototype → reliable, measurable production impact - Strong experimental judgment: formulating hypotheses, establishing useful baselines, designing evaluations, and interpreting results carefully - (Bonus) Prior experience utilizing and improving speech-to-text (STT), text-to-speech (TTS) models, and/or duplex models A PhD is not required. We value demonstrated research ability and the quality of the models, experiments, and systems you have helped build. BENEFITS & PERKS FOR ASSORTIES - 💸 Competitive Compensation – Including salary and employee stock options so you share in our success. - 📚 Lifelong Learning – Annual budget for professional development, plus training opportunities to help you grow. - 💻 Office Setup Stipend – We’ll outfit your in-office workspace so comfy as it's productive. - 🩺 Top-Tier Health Coverage – Medical, dental, and vision insurance, because your health comes first. - 🏖 Flexible PTO – We trust you to take the time you need to recharge and come back ready to crush it. - 🥗 Meals & Snacks – Lunch, dinner, and snack breaks that fuel great ideas. - 💪 Fitness Stipend – Your wellness matters. We reimburse monthly membership costs to support your health. - 🚆 Commuter Benefits – We cover eligible transportation costs to make your trip to work easier. - 👵 401(k) – Build your retirement savings. HOW WE WORK & WHAT WE VALUE Our team at Assort Health moves fast, stays focused, and is fueled by a desire to serve our customers and patients. Our company values guide how we work—they are present in how we show up, make decisions and work together to move our mission forward. We bring a Day One Drive, relentlessly striving to improve, keep a 5-Star Focus, as our customers are our lifeblood, always Answer the Call, remembering that ownership and accountability are paramount, and show up with One Pulse, because we are one team, with one rhythm and one result. Our team is growing and we are looking for motivated, hardworking, and passionate talent. If you want to make healthcare accessible for everyone, we’d love to hear from you! Please note: the Assort Health Talent Team will only email you from an assorthealth.com assorthealth.com email address.

Requirements

Must-haves

Healthcare spends $1.1 trillion a year on administration instead of care. If you want to help change that, apply below.
Lead research and engineering efforts to improve core conversational capabilities in production, including instruction following, tool calling, retrieval, and memory
Investigate research questions in speech and audio, and design experiments to test hypotheses about model behavior and capabilities
Build and iterate on end-to-end models and pipelines that optimize for quality, efficiency, and user experience
Identify opportunities for model-task optimization to ensure the right model architectures are used for different tasks based on performance, latency, and cost
Partner with platform and product engineers to integrate new models into production systems
Break down ambiguous research ideas into clear, iterative milestones and roadmaps
Train, fine-tune, validate, and further develop models built in-house at Hume and models developed by our customers
Design evaluation protocols and benchmarks that measure model quality, generalization, robustness, and progress against research objectives
Analyze datasets, model outputs, and failure cases; compare training approaches and use the evidence to guide the next round of model development
5+ years of experience with a minimum of 3+ in AI/ML engineering or research
Prior experience post-training and deploying LLMs in production environments
Fluency in Python and modern ML tooling (training, evaluation, inference, data pipelines)
Track record of taking research ideas from prototype → reliable, measurable production impact
Strong experimental judgment: formulating hypotheses, establishing useful baselines, designing evaluations, and interpreting results carefully
A PhD is not required. We value demonstrated research ability and the quality of the models, experiments, and systems you have helped build.
Lifelong Learning – Annual budget for professional development, plus training opportunities to help you grow.
Office Setup Stipend – We’ll outfit your in-office workspace so comfy as it's productive.
Flexible PTO – We trust you to take the time you need to recharge and come back ready to crush it.
Meals & Snacks – Lunch, dinner, and snack breaks that fuel great ideas.
Fitness Stipend – Your wellness matters. We reimburse monthly membership costs to support your health.
Please note: the Assort Health Talent Team will only email you from an assorthealth.com http://assorthealth.com email address.

Nice-to-haves

(Bonus) Prior experience utilizing and improving speech-to-text (STT), text-to-speech (TTS) models, and/or duplex models

Application prep

Use the role’s language and application details while tailoring your materials.

Keywords to mirror

aimlpython

Platform: ashby
Effort: quick

Benefits

Benefits detected in the listing.

Dental

Vision

401(K)

Pto

Wellness

Professional Development