Tech | Visa | Scholarship/School | Info Place

These AI startups stand out in Y Combinator’s Winter 2024 batch

Despite an overall decline in startup investment, funding for artificial intelligence has grown significantly over the past year. Capitalization in generative AI companies alone will increase nearly eightfold from 2022 to 2023, reaching $25.2 billion by the end of December.

So it’s no surprise that AI startups dominated Y Combinator’s 2024 Winter Demo Days.

Y Combinator’s winter 2024 batch has 86 artificial intelligence startups, nearly double the winter 2023 batch and nearly three times the winter 2021 batch, according to YC’s official startup directory. Call it a bubble or over-hype, but it’s clear that AI is the technology of the moment.

As we did last year, we took a look at the latest Y Combinator cohort (the one showcased at this week’s Demo Day) and picked out some of the more interesting AI startups. Everyone was eliminated for different reasons. But at a baseline, whether it’s technology, addressable market or founder background, they stand out from the rest.


August Chen (ex-Palantir) and Elton Lossner (ex-Boston Consulting Group) assert that the government contracting process is completely broken.

Contracts are posted to thousands of different websites and may contain hundreds of pages of overlapping provisions. (The U.S. federal government alone signs more than 11 million contracts each year.) Responding to these bids requires the involvement of entire business units, with support from outside consultants and law firms.

Chen and Lossner’s solution is to use artificial intelligence to automate government contract discovery, drafting and compliance processes. The two met in college and called her “Hazel.”


Image Source: hazel

Using Hazel, users can match potential contracts, generate draft responses based on RFPs and their company information, create to-do lists and automatically run compliance checks.

Considering that AI is prone to hallucinations, I’m a little skeptical that the responses and checks generated by Hazel are always accurate. But if they’re even close, they could save a lot of time and effort, giving smaller companies a chance to win the hundreds of billions of dollars worth of government contracts awarded each year.


Home nurses handle a lot of paperwork. Zha Tiantian knows this well—she previously worked at Verily, the life sciences unit of Google parent company Alphabet, and participated in a series of moonshot projects ranging from personalized medicine to reducing mosquito-borne diseases.

Documentation is a major time-consuming task for home nurses during the course of their work. It’s a widespread problem—according to one study, nurses spend more than a third of their time documenting, which reduces time spent on patient care and contributes to burnout.

To help reduce the documentation burden on nurses, Cha co-founded Andy AI with former Apple engineer Max Akhterov. Andy is essentially an AI-powered scribe, capturing and transcribing verbal details of patient visits and generating electronic health records.


Image Source: Aiandi

As with any AI-powered transcription tool, there’s a risk of bias—that the tool won’t work well for some nurses and patients, depending on their accent and word choice. And, from a competitive standpoint, Andy isn’t the first to do this. Marketplace – Competitors include DeepScribe, Heidi Health, Nabla, and Amazon’s AWS HealthScribe.

But as healthcare moves increasingly into the home, demand for applications like Andy AI looks set to increase.


If your experience with weather apps is anything like this reporter’s, you’ve been caught in a storm after blindly trusting a forecast of sunny blue skies.

But it doesn’t have to be this way.

At least, that’s the premise of Precip, an AI weather forecasting platform. Jesse Vollmar came up with the idea after founding FarmLogs, a startup that sells crop management software. He teamed up with Sam Pierce Lolla and Michael Asher, former chief data scientist at FarmLogs, to make Precip a reality.


Image Source: Prep

Precip provides precipitation analysis, such as estimating how much rainfall has fallen over a given geographic area over the past few hours to days. Vollmar claims Precip can generate “highly accurate” indicators accurate to one kilometer (or two kilometers) for any location in the United States, predicting conditions up to seven days into the future.

So what is the value of precipitation indicators and alerts? Farmers can use them to track crop growth, construction crews can refer to them to schedule crews, and utility companies can use them to predict service outages, Vollmar said. Vollmar said one transportation customer checks Precip daily to avoid poor driving conditions.

Of course, there is no shortage of weather forecast apps. But AI like Precip promises to make predictions more accurate—if AI is indeed worth the money.


Claire Wiley launched a couples coaching program while she was an MBA student at Wharton. The experience led her to study a more advanced approach to relationships and therapy, ultimately giving rise to Maya.

Maia, a company Wiley co-founded with former Google research scientist Ralph Ma, aims to help couples build stronger relationships through guidance from artificial intelligence. In Maia’s Android and iOS apps, couples can message each other in group chats and answer daily questions, such as challenges they think they need to overcome, past pain points and a list of things they’re grateful for.


Image Source: maya

Maia plans to make money by charging for premium features, such as therapist-created programs and unlimited messaging. (Maya typically limits texting between partners—a frustratingly arbitrary limit if you ask me, but it is what it is.)

Willy and Ma, who both come from divorced families, said they worked with a relationship expert to create Maia’s experience. The questions on my mind, though, are: (1) how sound is Maia’s relationship science and (2) can it stand out in the incredibly crowded field of couples apps? We’ll have to wait and see.

data curve

The core AI models of generative AI applications like ChatGPT are trained on huge datasets, a mixture of public and proprietary data from around the web, including e-books, social media posts, and personal blogs. But some of this data is legally and ethically questionable—not to mention flawed in other ways.

If you ask Serena Ge and Charley Lee, a distinct lack of data management is the problem.

Ge and Lee co-founded Datacurve to provide “expert-quality” data for training generative AI models. It’s specialized code data that Ge and Lee say are difficult to obtain due to the expertise required to label it for AI training and a restricted use license.

data curve

Image Source: data curve

Datacurve has a gamified annotation platform that pays engineers to solve coding challenges, which contributes to Datacurve’s training datasets for sale. Ge and Lee said that these data sets can be used to train models for code optimization, code generation, debugging, UI design, etc.

It’s an interesting idea, to be sure. But Datacurve’s success will depend on how well its data sets are curated and whether it can incentivize enough developers to continue building and improving them.

#startups #stand #Combinators #Winter #batch

Leave a Reply

Your email address will not be published. Required fields are marked *

Table of Contents