Chris Lovejoy, MD (@ChrisLovejoy_) on X (original) (raw)

Member of Technical Staff

@anthropicAI

Prev: medical doctor, ML engineer, founder

• Pinned
  
  My hot take: The vertical AI products that will win are not the ones with the most sophisticated models or techniques - but those with the best system for incorporating domain expertise. The LLMs in the product need to understand the industry-specific and customer-specific way
• 
• 
  I'm thinking about writing a short eBook for people from medical backgrounds wanting to get into machine learning or data science. If this sounds like you, what questions would you want answered?
• 
  Building a dashboard for domain experts to review AI outputs is critical for domain-specific LLM apps. Principles we followed at
  @AnteriorAI
  when building ours: (1) Optimise for providing all relevant context as intuitively as possible Our doctors and nurses say the hardest
• 
  AI and Machine Learning is hard. But it’s the future of Medicine. You can learn it and scale your positive health impact. Here’s how: (1/5)
• 
  Applying AI to medicine is hot right now. But how can we do it right? Two landmark papers were recently published, to tell us how:
• 
  Making an LLM product MVP is easy -- building a product that works at scale is hard. Last week for the
  @aiDotEngineer
  conference I shared our journey at
  @AnteriorAI
  from prototype to production, now serving health insurers covering >50M Americans. Our special sauce? Real-time
• 
  I’m looking for “user-friendly”, open-access medical datasets. I am designing a series of “Coding for Medicine” tutorials. Aimed at medics learning to code + computer scientists interested in healthcare. Ideally, data is pretty clean and easy to access. Any suggestions?
• 
  Turns out I've hit 1,000 subscribers on my YouTube channel (I make videos about machine learning, coding and healthcare) That's lot of people, although still small by YT standards (Gotta start somewhere) Do you have any questions you'd like me to answer in my 1000 subs video?
• 
  I built an MCP server for trusted medical advice You can ask it medical questions on any topic It anchors its answers in up-to-date medical information, only from reputable sources It's fully open-source, self-hosted, and doesn't rely on third-party APIs (Important caveat:
• 
  I'm thinking about doing short blog series / email course on 'how to read a medical AI paper'. Initial idea is to cover: - deciding papers to read - a process for extracting key info - dealing with new terminology - combining insights across diff papers Would you be interested?
• 
  Our latest study in the BMJ shows that research comparing #AI to doctors often have over-stated conclusions, and use methodologies with high risk of bias.
  @bmj_latest
• 
  Super excited to come out of stealth as part of this team! Watch this space- and if you want to be a part of it: co-helm.com/p/careers
• 
  Thinking about making a medical MCP server Initial ideas: - medical codes: receives a medical condition name, returns the most relevant code (ICD10, SNOMED-CT) - medical guidelines: receives a medical condition name, returns the treatment protocol - medical research finder: