3 ms·

AuxHealth | Founding ML Engineer | Remote / Carlsbad, CA | US Only | FULL TIME | https://auxhealth.io https://auxhealth.io We are building generative AI system

by ivalm 3y ago

AuxHealth | Founding ML Engineer | Remote / Carlsbad, CA | US Only | FULL TIME | https://auxhealth.io https://auxhealth.io We are building generative AI systems that facilitate clinical workflows by interviewing patients, writing notes for their clinicians, and performing chronic disease management. We aim to allow clinical teams to offload clerical tasks so that they can care for patients at the top of their license. We've just raised a pre-seed round ($2M) and are looking to hire an MLE. The team right now consists of 2 engineers and a medical SME. We're looking for someone with: * High level of proficiency in deep learning frameworks (e.g., PyTorch) and language model libraries (e.g., Hugging Face Transformers) * Research or industry experience with transformer-based language models * Expert level of knowledge of Python Bonus points for: * Experience in fine-tuning multi-billion parameter models * Experience in large model deployment and efficient inference optimization * Experience with Typescript/React * Experience in the healthcare domain Interested? Email: ilya@auxhealth.io

reaperman 3y ago
I am generally very cynical about most AI startups but I think there's tons of room in this space -- it's an obvious fit, can provide simple workflows to solve pain points for both providers (your customers) and patients. Your model also builds inroads towards tons of extensible real-world applications in the healthcare world. Worth noting Ambience Healthcare (and probably many others) approaching this space from a slightly different angle, while you're making the pre-intake forms more useful, they're transcribing the doctor-patient conversations. Overall I think both approaches are solid. Be careful though as there's an increase in dark patterns around intake forms in the past couple years, I've noticed some concerning privacy practices. As a patient, this centers around being handed an iPad to fill out which contains an agreement that gives away a lot of the privacy of your health data. I expect more patients to start complaining about this practice and for it to eventually be clamped down on by some federal authority. News articles have noted these privacy concerns around Phreezia[0], BetterHealth, and GoodRx so far. I doubt your group is headed that direction though -- most of the companies / business units who do that have a very limited business model that stops at "gather health data and either resell the data or use it internally to target advertisements", whereas your business appears to be strongly focused on creating real value for doctors and patients. https://archive.ph/dzx7u https://archive.ph/dzx7u

by ivalm 3y ago

reaperman 3y ago

I am generally very cynical about most AI startups but I think there's tons of room in this space -- it's an obvious fit, can provide simple workflows to solve pain points for both providers (your customers) and patients. Your model also builds inroads towards tons of extensible real-world applications in the healthcare world. Worth noting Ambience Healthcare (and probably many others) approaching this space from a slightly different angle, while you're making the pre-intake forms more useful, they're transcribing the doctor-patient conversations. Overall I think both approaches are solid. Be careful though as there's an increase in dark patterns around intake forms in the past couple years, I've noticed some concerning privacy practices. As a patient, this centers around being handed an iPad to fill out which contains an agreement that gives away a lot of the privacy of your health data. I expect more patients to start complaining about this practice and for it to eventually be clamped down on by some federal authority. News articles have noted these privacy concerns around Phreezia[0], BetterHealth, and GoodRx so far. I doubt your group is headed that direction though -- most of the companies / business units who do that have a very limited business model that stops at "gather health data and either resell the data or use it internally to target advertisements", whereas your business appears to be strongly focused on creating real value for doctors and patients. https://archive.ph/dzx7u https://archive.ph/dzx7u