SignalFire

This is what a modern venture capital firm looks like

Heading to #TechCrunchDisrupt2024 next month? Don't miss SignalFire's own Heather Doshay's session on hiring in 2025!

Celebrating our SignalFire portfolio companies chosen by Business Insider as the top 22 most promising healthcare AI startups: Health Note, CodaMetrix, and Grow Therapy!🎉 Congrats to Joshua Reischer, MD, Aaron Rau, Gregg Boyle, Hamid Tabatabaie, Jake Cooper, Alan Ni, Manoj Kanagaraj, MD! We're excited for your journey ahead.👍 Due to overwhelming founder interest after our event featuring Grow Therapy, SignalFire is back with our next #AI Lab event to unpack Moxie's playbook with Dan Friedman! Dan will share more about Moxie's founding story and lessons learned from building a business-in-a-box platform for medspas this time. Feel free to apply to join this event on 9/10, our founder round table dinner in NY on 9/11, and watch the Grow event recording if you missed it below 👇 #HealthTech #BizinaBox

We're on the cusp of a radical transformation of healthcare by AI, led by startups like Grow Therapy's tools for therapists, CodaMetrix's automated insurance bill coding, and Health Note's patient intake platform.

NY-based builders, join Sooah Cho next week for an exciting discussion on how #genai is transforming pharma! Reg link in comments:

How do you build a rapidly scaling healthtech business like Grow Therapy by providing a "business-in-a-box"? - Focus on solving one stakeholder's problems first to build trust and expand from there - Prioritize ruthlessly to maintain service quality while scaling quickly - Leverage technology to improve care delivery and provider efficiency - Build strong payor relationships by addressing access, quality, and triage needs Check out Sooah Cho and Wayne Hu's deep dive on Grow Therapy's playbook in our latest blog:

“Anecdotes about the decline of tech in San Francisco are overstated," we told TechCrunch. It wrote about our SF Is Back report that found that The Bay Area is home to more top startup employees than Seattle, Austin, and NYC combined. "SF still dominates all other U.S. cities when it comes to concentrations of tech talent and capital, and its lead is even larger when it comes to the recent AI boom [with 38% of early-stage AI rounds]." Far from dead, SF remains the best place to build. Check out the TC article and our full SF Is Back report in the comments.

"Lago moved to SF instead of New York" - Why? I've been asked this so many times for the past months: by founders, team members, candidates, investors. - As always, I thought "I should write about it" (it's a more scalable way to answer the question). An even better opp was to chat with the fantastic Julie Bort at TechCrunch (Thanks Josh Constine!). The TLDR: SF is the best place to expand Lago: from a customer pool and talent perspective, *despite* the timezone. Before having any physical presence in the US, we already had 30% of our user base here, and we moved in the Bay Area to amplify this. SF is also a place where "luck is manufactured" or, in 🇫🇷 we'd say "provoquer sa chance"! We've already made our first hires in SF in a few weeks and are still hiring! (ping Alison Eastaway ;) ) Immensely grateful to the amazing local support group here in the Bay Area! Oana O. AJ Solimine Nicolas Dessaigne Diana Hu Romain Huet Cathy Han Aishwarya (AG) Goel, and of course to my partner in crime Raffi Sarkissian! PS: to answer another of the most common questions: Deel Alex Bouaziz Shuo Wang ahmed mousa 👋 helped me for the visa, ping them directly! 🙃

SF really is back. SignalFire’s Beacon data platform reveals San Francisco leads the country as the #1 place to found, fund, and hire for startups. More stats include: -The SF Bay Area is home to 27% of all startup engineers and 49% of all big tech engineers.  -SF leads with 26% of all top VC-backed early-stage funding rounds, and 38% of AI funding rounds.  -Despite the “SF is dead” narrative, its share of founders and startups both dropped less than 10% in the COVID era Though we love our NYC office too, we're proud to be headquartered in SF and contribute to its lively startup culture. Check out our full SF Is Back report in the comments for more stats and graphs from our head of research Asher Bantock‎‎.

Four trends to watch in one of the most exciting areas of AI—called reinforcement learning—from leading experts: Our partner and machine learning scientist Oana O. moderated a panel at Ai4 in Las Vegas with William Viet Tran (senior engineering manager at Pager Health), Jasleen Singh (senior principal research scientist at Dell), Alfredo Garcia (Texas A&M professor), and Sainyam Galhotra (Cornell assistant professor). Here’s what we learned about reinforcement learning (RL), a branch of machine learning where an agent learns to make decisions by interacting with an environment (think driverless cars): 𝗖𝗼𝗻𝘁𝗲𝘅𝘁𝘂𝗮𝗹 𝗿𝗲𝗶𝗻𝗳𝗼𝗿𝗰𝗲𝗺𝗲𝗻𝘁 𝗹𝗲𝗮𝗿𝗻𝗶𝗻𝗴: An autonomous car can handle a closed course well but struggles when a family of ducks tries to cross the road. This strategy takes the environment into account when training, which makes training much more difficult but is often necessary to move from theory to reality. 𝗦𝗮𝗳𝗲𝘁𝘆: RL can come up with custom treatment plans for patients based on their history or discover new drugs, but are its recommendations safe? For a lot of applications, the “do no harm” principle will have to apply. 𝗥𝗟 𝗳𝗿𝗼𝗺 𝗵𝘂𝗺𝗮𝗻 𝗳𝗲𝗲𝗱𝗯𝗮𝗰𝗸: Ever read something and think, “this sounds like it was written by AI?” RLHF focuses on pushing AI systems to become more human, with an expert (human) comparing the AI’s responses to those of a human and tweaking the model. 𝗜𝗻𝘃𝗲𝗿𝘀𝗲 𝗿𝗲𝗶𝗻𝗳𝗼𝗿𝗰𝗲𝗺𝗲𝗻𝘁 𝗹𝗲𝗮𝗿𝗻𝗶𝗻𝗴: With IRL, the AI figures out what to do based on what it observes instead of being pre-preprogrammed. In the driverless car example, researchers could input reams of real driver data and let the car figure out what is important, instead of inputting countless commands like “go at green light” and “stop for pedestrians” and letting the car figure out what to do when both conditions are present. For more details, see our latest blog post (link in comments).