E5: Reviewing startup pitches from Punk Rocker Neuroscientists to Data Scientists | YC W24

Welcome to the Superconnector podcast. I am Matt Joseph, your host. We've got another round of startups to look through today from the YC winter 2024 batch. First up we have Meticulate. Effortless business research. Okay, let's jump into this and see what we got.

Meticulate provides finance professionals, easy access to world class research. Okay. Makes sense. Good business research is slow, costly, and inaccessible. And so we've got a little preview here of how the product works. Meticulate uses LLM pipelines to emulate an analyst research process.

Okay. So analyst research is a massive business. All the biggest investment firms have analysts who do this kind of work. So I know this to be a very valuable and lucrative field to be in.

I think we should get a little bit more context about who these guys are.

Wilbur and Joseph have been close friends for 15 years after growing up two blocks away from each other in Houston, Texas. Alright, So we got some neighbors. Wilbur is a prior McKinsey consultant who suffered through hundreds of hours of business research. And Joseph is a prior startup engineer and NLP specialist.

We've got a really nice high res photo of them standing outside of YC's office in mountain view.

So I really liked this business. I think it's really smart. I've worked on teams that were analyzing businesses.

I did that at BCG a little bit. And it was an incredibly time intensive process.

So that makes a lot of sense. I mean, it's not clear what their traction is from this launch. But they have the right experience. Consulting is kind of shitting the bed right now. Makes a lot of sense. Wilbur and Joseph. Good luck.

Next up. Infinity business banking for cross border SMBs and startups.

Infinity simplifies banking for global businesses and enables cross border Indian companies to conduct transactions and receive payments in local currencies.

So we have an Indian company that is going after cross border payments and finances. It's not clear if they're actually building a bank or if this is just a transactional system. There have been some very successful businesses like this in Europe. Sounds like they're taking that same thinking and applying it to the Indian market.

Okay, let's jump down to the problem.

In general, for every dollar a business moves from the U. S. to India, business loses 7%.

Okay. So. These are not the first guys to come along and say, they're going to do cross border payments in India. India is an exploding venture market. Indian payments in particular is an area where there have been a lot of founders who are focused on it. But I'm curious to see if the team actually has some unique experiences that they bring to the table, because I think that could actually make a big difference here in terms of getting them up and running.

So let's scroll down here.

Surav and Siddhartha are siblings. This is their second startup together.Sourav was the head of growth and product team at Paytm money. He also successfully launched India's first FD banked credit card. Siddharth is a finance and tech geek. He spearheaded the in app purchase module for the Amazon app store utilized by app developers to monetize digital content globally.

So this is a really strong team. I love when you have teams that are siblings, because you know that their bond is deeper than the business. There are outliers to that but in most cases, you know, that their relationship is going to stay strong. You're probably not going to have a cofounder breakup when you have siblings who are working together. They also have great experience in the financial services business. Paytm money is India's largest wealth management platform.

Both of them have experience there and we have another Amazon alum which overall I always liked that and I always think it's great. So like the team. Like the problem, it's highly competitive. But. I think these guys might be the team to pull it off. So Sourav, Siddharth congratulations on launching and good luck.

Next up, we have tile. AI first SQL notebook for building stunning data apps.

Okay, so this is an interesting technical challenge here.

Let's learn about the founders. Jenica and Jessica have collectively spent nine years data wrangling at Palantir. Before Palantir, Jenica studied math and CS at Harvard and Jessica studied CS at MIT.

So we have a very technical team working on a problem that it seems like they encountered at Palantir. I think that's actually a really great base for building. Palantir has some of the most interesting projects of any company. And they've done a great job as a public company too. So I like that background.

And we have a video here. Let's play this and see what they say.

how Dwight from the office.

would use Tile to make Dunder Mifflin's sales decisions. So Dwight noticed that their map paper profits for this month have made up a huge portion of their overall profits, and he wants to understand why. He can go ahead and select last month's transactions data to Take a look into.

what's going on.

So the UI here is pretty slick. It looks like a lay person would be able to use this, even if they didn't know SQL. And I think that's actually a really compelling value prop. Essentially the way I would think of this is that you have people who are non technical or lightly technical who want to be able to dive into data analytics using SQL and tile makes that a lot easier. That's what I take away from this. I think that's actually a really smart system. I could see it expanding into a very significant analytics suite. That seems like a great idea.

So this is a good business and we have a nice photo of these guys. It looks like they're actually at the new Y Combinator office in San Francisco smiling. I like that they look happy. Congratulations on launching tile and I wish you guys well.

Okay. Up next is Speck. AI powered web automation.

All right. customers use Speck to extract data, perform LinkedIn outreach, collect the newest research papers, gather data about social media groups and more.

Okay. So automating web tasks.

This feels like a tool that does web scraping. I've used a few of those in the past. But it seems like this one is actually a lot more robust than some of the web scrapers I've used. If they're doing outbound and taking care of other things that you might do on the internet. I think that's actually really interesting and valuable.

Our browser extension allows you to record workflows just by doing them. Speck automatically translates your workflows into repeatable steps, making it simple for anyone to create web based automations and data extraction processes.

So that's actually really interesting. I think for many web tasks like this. It can end up being dynamic, like the way you do it one time is not going to be the way you do it another time, but it looks like they've actually taken that into account. So here's what they say.

By keeping track of intent, the browser automations will repair themselves automatically when something goes wrong. Since pivoting two weeks ago and launching our beta, we've gotten our extension published on the Chrome web store and onboarded our first paying customers. And so they have gifs of the workflow. How this exactly would work. They're actually pulling data off of this website that we're on right now.

The YC launch data.

and Let's learn about the team.

Raghav is an ex Anduril, an ex Amazon engineer. That's great.

Lucas is ex Roblox. He's also an engineer. And he built multilingual auto-complete for 160 million users. He's the founding engineer of sweep, an autonomous LLM coding agent with 7, 000 stars. Stars is like a way that developers products.

Okay. So we've got a really strong team here. I love their backgrounds. Seems like they are good friends. One yellow flag is that this is a recent pivot. I would want to dive in and understand what they were doing before that led them to this.

And they've said they onboarded paying customers, not quite clear who those customers are and whether there's a repeatable motion to acquire those customers. But the technology seems to be really cool. And I think it's a smart idea.

So A lot of times, instead of pivoting, people would just shut their companies down. And I actually give these guys a ton of credit for executing a pivot.Up next. Pretzel. Open source and AI native replacement for Jupyter notebooks. Well, I love pretzel as a brand, so I'm excited to jump into this and see what we got.

we're building the modern Jupyter notebook that fixes many longstanding problems with Jupyter. Pretzel improves on Jupyter with native support for AI tools like GitHub co-pilot and LLMs built in collaboration.

Best of all, we are fully open source. As our first product, we built just one feature pretzel. A visual data exploration tool that runs locally in the browser. See our demo below. Okay. So we've got a nice little GIF here. Outlining exactly how this works.

I love the way that they have framed this they're going after Jupyter's customers. So it's very, very clear who the user is for this and why they would want to switch. I think that's an incredibly compelling base for a business.

I also really like how they've branded it. So they're basically saying to Jupyter notebooks customers. Hey. We got a better Jupyter notebook. I just love that. I think it's really smart.

So let's hear what they have to say about Jupyter.

Jupyter notebooks started in 2011 as a nicer, interactive way to work in Python. It's currently used by millions of data scientists and analysts. Despite its popularity there are many long standing problems with Jupyter. So they're going after a number of different problems, but I like how focused they are because they said they're only releasing one feature to start. So they're signaling what the vision is. But they're starting with something very focused and very specific, which clearly they think is the biggest segment of the pie. Or the most accessible segment of the pie. And I think that's really smart.

And we've got a nice photo of them hanging out on the beach together. They are saluting the camera on the beach. I love these guys. One of them has a Campbell's soup shirt on. I love it. I love it. Ramon. Great shirt.

Prasoon and Ramon. Love the salute. Congratulations on launching.

All right. Next up Senso. AI powered knowledge base designed for credit unions. Ooh, credit unions. I spent a long time looking at credit unions back in the day. Let's jump into this.

Senso is building an AI powered knowledge base to expand customer relationships, starting in the credit union space.

credit unions are for many people a cheaper alternative to using banks. Let's learn about the team. We're Saroop and Tom.Over the past decade, we've been developing AI solutions for banks and credit unions, which provide marketing and support teams with insights to boost customer retention rates. And we've got a couple of photos here. We got one photo with them hanging out in the forest, it looks like they just went hiking. And another photo with what looks like family. So I'm assuming that both of them are married.

It looks like we've got a couple of kids in the mix. That's awesome.

we have a customer testimonial. First, one of these I've seen in a while. I love customer testimonials. We have Lisa Marie Meyer, SVP director of Minnesota branches, true stone credit union. Quote, Senso is transforming how we handle member queries. Our pilot is proving that our staff can now provide answers within seconds, not minutes, a game changer for member satisfaction. Team members are requesting to join the pilot. And we are excited to launch Senso to all of Truestone.

Well, that is a very successful pilot. Well done. We have another one. From Gary Jeter, EVP and CTO of Truestone credit union. So the same. Credit union. The continuous improvement Senso brings to our documentation has been invaluable. We're not just keeping up, we're setting the pace for operational excellence.

Okay. So, this is actually really good.

It's what some people would call a boring business. I find it to be very interesting. Financial services as a whole. And if we talk just FinTech more generally is highly, highly saturated. But I actually think these guys have found a pretty interesting and unique angle, and I love the way that they are customer centric. So few YC startups put testimonials into their launches. I think all of them should. Testimonials bring these things to life. These guys did a great job with it. Seems like they have a great background. I love the team. This is a good business. Up next Risotto. Turn any slack channel into an automated IT help desk in 60 seconds. Well, I'm getting hungry, just seeing what their company name is so I assume I'm going to be really excited to check out the business.

Risotto automatically solves IT support tickets using AI. We help AI teams improve their resolution times, and it support all directly in slack. And so this is a slack bot. It looks pretty slick. Everyone uses slack now. So I think this could be interesting.

Let's meet the team. Aaron, Alex, and Chris met almost 10 years ago at HelloSign, which was YC 2011, as early employees. Alex led the HelloSign it team through the acquisition by Dropbox. So HelloSign is one of the great YC stories becauseThey were essentially building like a DocuSign competitor and they ended up selling to Dropbox, which was another YC company. And it was a great outcome. I mean, it was in the hundreds of millions of dollars that they sold.

It sits in YC lore because you have two YC companies, you have Dropbox, which, you know, one of the great successes of YC, and you have HelloSign, which got a great outcome. So that's like YC feeding YC. So cool little story within the network.

All right. Let's talk about the problem. IT teams spend too much time manually provisioning software and answering the same question repeatedly. One of our co founders, Alex saw this play out constantly while running it at Gusto and Dropbox, which inspired us to do something about it. Alright, So we've got a video here. Let's play this.

Risotto can turn any slack channel into your it help desk in just 60 seconds. Let me show you how easy it it is.

Head to our integration tab, Click Add to slack. Select the. Environment we want to add Risotto to in this case. Risotto demo. Select the permissions. Then Risotto will send us a DM and ask who do we want to escalate when it doesn't know an answer to a question?

I like this a lot. The differentiator for this, as opposed to another help desk solution you might use is that it's deeply integrated with slack. So you can add it to a place where your team is already hanging out. That makes signing up getting going really easy and really fast. I think that's actually a huge feature.

It's going to appeal a lot to early stage startups that are trying to offer this. Mid stage startups as well. I think at the point where somebody has a lot of infrastructure around their it help desk, it's a little bit of a more difficult sale to make. But I think for smaller companies, particularly technology companies who are all over slack, this is a really, really smart strategy. Just very easy to use.

Risotto's benefits. Risotto improves the key metrics that it teams care about such as time to resolution, time to first response and employee satisfaction ratings.

So, this is cool. I see two things that are great strengths here. Number one, the team has a lot of great experience in startups. These are deep YC founders. Multiple YC startups. That's great. They know how to build stuff. The second thing is it's a very easy to implement solution.

This is a problem that has a ton of competitors floating around. So having something that people can easily sign up and try and get comfortable with? Incredibly valuable and the deep integration with slack also really helpful.

There is a challenge here, which comes up a lot when companies are so heavily dependent on a particular interface. Yeah. You don't know what the big company is going to do. Slack or its parent company Salesforce could at any time decide, we want this use case and we're going to shut these guys down. That's an existential risk for this company. So over time, I imagine that they're going to get into other environments, but this is a phenomenal wedge while it's available. Guys. Congratulations on launching.

Too bad there's no food involved in this, but we can let that one slide.

Up next Piramidal foundation model for the brain. Get instant diagnosis for the vast majority of neurological conditions using our foundational AI. Well, now this is interesting.

We are building a foundation model from the ground up, trained on massive and diverse Corpus of brainwave data. In other words, similar to how large language models are trained on human language, our model is trained on brain language.

Crucially similar to how ChatGPT can be fine tuned to become a lawyer or a Python developer, our brain foundation model will easily be fine tuned to become a neurologist co pilot or a biomarker discovery assistant.

Wow.

I think they may have just broken my brain with this one. This is incredible. Essentially they are using brain waves to create a model that will allow people who work with the brain to do things the same way that people can do things with ChatGPT.

Yeah, this is really, really smart.

So let's jump into this a little bit further. Our brain foundation model will address a variety of use cases with minimal training, such as diagnosis of a vast majority of neurological conditions, drug and biomarker discovery in clinical trials, remote monitoring of brainwave abnormalities, real time tracking of brain health such as cognition or head injuries, empowering devices such as apple vision pro and Meta Quest to translate brain signals into everyday actions.

Wow. Okay. So let's talk about the team. During his Ph. D. at London's St. Thomas EEG department, Dimitris discovered a key insight. EEG is a crucial test for neurological disorders. There's a substantial rise of EEG data being collected due to the deployment of remote EEG devices.

Despite these remote devices, continuously recording patients at home. there's no way to review this enormous and continuous stream of incoming data. Conversely There is an extreme shortage of neurologists leading to wait times lasting several months.

Okay. So these are. Really great problems. if the idea is that they can take all this data, which isn't really being used anyway and build an incredible model that actually helps them to diagnose more patients and helps drug companies to do this biomarker analysis. I think that would be incredible.

All right. so.

More on the team. Dimitris has a PhD in neuroscience and AI, and has been working on EEG for 12 years.

Chris has a product background, building healthcare and AI products across Google and Spotify. We are both avid musicians and love punk rock. Well, I am surprised by the punk rock reference here, but I kind of love that.

I'm imagining two guys with those spiked neck chains and spiked hair, just rocking out.

But during the day they put on lab coats and, you know, go into the office.

I love this team. I think this is a fascinating problem. I don't know enough about the brain to be able to assess whether or not this makes sense. What I do know from the few neuroscientists that I've talked to is that the brain is still somewhat of a black box. And it seems to be one of the areas that technology can actually make a massive difference because we ourselves still know very little about how it really works.

So. This could be a groundbreaking technology. This could really change people's lives. Particularly in the beginning, people who have neurological problems. I imagine it's gonna take them a long time to roll this technology out, but this is a fascinating company. Topo AI sales agent custom trained for each company to reach out to the best leads.

All right. Let's jump into this.

The agent is trained to handle all outbound tasks, automatically finding the best leads in the industry specific channels, and then automatically contacting prospects with a personalized approach.

Okay. So we've got a lead gen company. There are lots of YC companies that have basically said we are going to get the best leads.

It seems like the difference here is they're saying we're not only going to find the leads, we're going to contact them on your behalf. So essentially they are co opting the entire sales process with their artificial intelligence solution.

I have the same general critique of all of these sales products, which is I don't want more fucking spam emails in my inbox. And. I want these companies to figure out a better way to sell.

I really, really hope that these guys have that because I get too many emails which are not context aware. They'll just email me with a sales pitch for something I have no need for. I got a sales pitch this morning for a marketing solution for FinTech businesses. It's like, why on earth would you send that to me? Having greater personalization is good.

This feels to me like a band aid on this bigger problem, which is that there are too many transactional emails floating around. I hope that they move the needle on this. I hope that they do well with it. my bias is generally against solutions that are going to send out an enormous amount of emails that will require huge training to get anywhere. But I'm open to hear why this is interesting.

Dan and Robin were together as sales and head of outbound at Aircall from 10 million to 120 million ARR. Leonard is a data scientist and AI expert and managed 10 people in his previous startup.

And we have a photo of them hanging out outside of the Y Combinator logo in the Mountain View office.

I hope that they are doing something which fundamentally makes the sales emails I get better, Dan, Robin, Leonard, congratulations on launching.

Up next, Trieve, Infrastructure for Search Teams. Looks like their name is built on Retrieve. I like that. Let's jump into this.

We want to power every search bar on the internet. Now that is ambitious. You have my attention. Trieve offers source available, all in one infrastructure for entire search teams.

All right, let's learn about the team. Denzel and I are software engineers who walked into this problem trying to build semantic search for a knowledge management system that helped business teams argue better. When we went to go sell the arguing app, we found out that the underlying search infrastructure was what people really wanted to buy.

Oh, it looks like they're on a podcast together It's like they're having a good time.

So there's another YC company that has created a dominant position in business search called Algolia. I'm sure these guys have run into them. It's not clear to me what makes this special.

Like, I don't love the founding story here. They were building a knowledge management system to help business teams argue better. That doesn't strike me as an issue I've ever seen on any business team that I've been on. And I've been on dozens at this point. don't fully understand the need for this.

I think these guys are young. I think they're just figuring it out. They are working on something which is a hugely ambitious problem. It's just not clear from their description exactly what they're building or how it's differentiated in the market. So I think they have some more work to do, but I like the ambition.

It's exactly the right sort of big North star. I hope they can get their product to match up with the ambition that they have. Up next RetailReady, automating supply chain compliance.

Okay. got to say, I love this photo of them they have right here up front. They are wearing construction vests, like the luminescent vests that the construction workers wear. That's great. We're Sarah and Elle and we're excited to launch RetailReady. Before we joined yc, we worked at a supply chain unicorn startup called Stord, where our jobs involved making their warehouse shipping facility run

efficiently. Let me explain what actually happens in the warehouse when you order something from Walmart. Every time workers in a warehouse box an order, they're supposed to reference 100 plus page manuals that include precise packing instructions, like where to put the shipping label. If retailers receive a noncompliant shipment, such as incorrect label placement, they issue brands a compliance violation fee called chargebacks. These mistakes result in hefty fines amounting to 40 billion annually. RetailReady's mobile application replaces the paper manuals, making it easy to stay compliant and mitigate these fines without any training time or integrations required.

All right. So this is one of those incredible nascent problems that you'd only know about if you actually worked in a warehouse. I love that it's two young women who are working on this, not your typical warehouse workers, I would say, but love that they have taken this problem on. And now they have examples from these guidelines.

Yeah, this is great.

It makes a lot of sense to me that this is a problem that not too many people know about. Anytime you have these big old entrenched brands, it's difficult to get them on board. It seems like they probably need both sides to agree to use this for it to work. Although you could just have the brands use it if it's only about replacing those guides with a mobile app, but clearly what they see is a platform sitting in between retailers and brands managing the shipping process as hands on as they can be with technology.

So that's cool. I like it. Love the story. Seems like a cool business.

And I absolutely love this photo of them in the warehouse. So Sarah and Elle, on launching.

Well, that's another round of startups in the books, another 11 great startups. Thanks for watching. Thanks for listening. I'll see you again soon.