AI Agents | YC F24 Startups
Which AI agent developer will be in demand in the next few months?
Hi!
What do you think is the most significant trend in AI? Lately, I've been considering AI agents as the main game changer. Tech companies and investors are already betting on this niche. Just so you understand the scale, AI agent startups raised over $8.2B in the last 12 months. And, it's expected that more will follow.
So, let's break down this topic. In this edition, we’ll cover these questions:
- Why Are AI Agents In Such High Demand?
- Who Are the Main Beneficiaries of this Niche?
- What Projects Should We Be Following Right Now?
Keep your mailbox updated with practical knowledge & key news from the AI industry!
AI Agent Opportunities
The first (and easiest) thing we must do is define AI agents. There are quite a few options here, but I prefer this one: AI agents are autonomous programs that can perform tasks without constant (or partial) human intervention. They can be represented as writers, lawyers, consultants, and other specialists in digital professions.
While Elon Musk and his colleagues are preparing the ground for introducing autonomous robots into our homes and offices, we have a great opportunity to lose AI in the online space. And it's pretty funny, because until a few years ago, we thought the automation would start with the real world.
You don't have to go far for examples. Salesforce has already started shipping its Agentforce, and Swedish Klarna is using AI to replace 700 employees. The above (and many others) plan to expand their AI agent deployment capabilities as LLMs evolve.
Why are they doing this? There are many reasons.
Here are the main ones:
- Autonomy: AI agents act independently, making decisions and acting on them without direct human control. Less human error is less human error.
- Purposefulness: Humans set common goals, and AI agents determine the best course of action to achieve those goals. AI also adapts more quickly to changing circumstances.
- Perception & learning: Even a good expert is too slow to learn and has limited ability to master new skills. New models progress much faster.
- Cost-effectiveness: The cost of implementing AI agents is fully offset by the fact that they don't need salaries, weekends, and vacations.
Inflation will also affect the employer only at the level of contributions from the company that maintains and optimizes the AI agent.
Sharing is caring! Refer someone who started a learning Journey in AI!
Winners In The Race for AI Agents
The question of who will benefit from the mass adoption of AI agents can be looked at from different angles. It could be consumers who will receive a higher level of service, businesses that will become more cost-effective, or companies that will monetize the new generation of bots. To avoid wasting time, let's focus on those already making money in this niche.
And strangely enough, big tech companies don't seem to be a good fit so far. Google, Microsoft, Salesforce and many others are indeed already delivering such solutions. However, it's still too early for direct cash flow. Unless OpenAI, which recently raised a record $6.6B funding round can be considered the leading beneficiary.
We've covered that topic here:
Still, the big winners in this race so far are small to mid-sized startups. According to PitchBook, more than 156 deals worth $8.2B have been closed in the AI agent space in the past 12 months. The median deal value grew 46.9% to roughly $3M.
So yes, startups in this niche are feeling just great. To be specific, analysts name the following sectors in which AI agents will be most in demand:
- Software Development
- Marketing
- Legal Services
- Education
- Customer service
Startups in these areas (e.g., coder support platform Cognition and legal AI developer Harvey) have already received a significant amount of investment and are expected to continue to attract new rounds.
But what kind of AI startups will join the list of in-demand developers with huge investors?
AI Agents Developers in YC F24
To answer this question, we looked at new participants in Y Combinator (which regularly supports AI startups) and found several suitable projects in the latest batch. What is interesting is that among them were not only developers of the agents themselves but also those who create the basis for providers of such solutions.
Let's break them down.
We have previously analyzed one of the YC cohorts:
Should we make this a regular feature?
HumanLayer | Human-in-the-Loop for AI Agents
HumanLayer is an API and SDK designed for AI agents that need to connect with people in Slack or via email. The developer provides an example of an AI agent who decides to send an email to a potential customer or update production data. In this case, the system will contact humans to get approval for the appropriate action.
Dexter Horthy is the founder and CEO of HumanLayer. He previously launched a project for catalytic developers that grew into an actual startup.
Bramble | AI-powered buyer real estate brokerage
Bramble is a startup that operates as an AI-powered real estate brokerage. The firm provides comprehensive home buying services for a flat fee. The developer promises to speed up the entire process, from organizing tours to closing deals. Bramble has already conducted testing in California and in a couple weeks has received hundreds of registrations, several tour buyers, and multiple offers across the state. And all of this, the company says, with almost zero marketing costs.
Bramble has three cofounders, Nimesh Chakravarthi, Daniel Hsu, and Deepan Mehta. All three are former LinkedIn employees with backgrounds in software development, customer support, marketing, and AI.
Sandra AI | Voice AI receptionist for car dealerships
Sandra is a developer that has focused on a specific scenario where AI agents are used. Namely, servicing car dealerships. Their AI handles the calls of any customers with support for different languages. The creators say that Sandra differentiates itself from its peers due to its deep integration with car dealership management systems, which allows the AI to handle even the most complex requests.
Sandra AI is another startup with three founders from the same large firm. Badr El Idrissi Mokdad, Ismail Majjad, and Skandere Shali are from McKinsey & Company.
Protegee | Payments API for AI agents
Protegee works at the intersection of fintech and AI agents. It is a startup that has developed a system that allows digital assistants to be trusted to accept payments. The developer takes care of compliance, security and fraud management, allowing the customer to focus on creating a quality experience. Web and agent-to-agent transactions will be added to phone payment processing in the future. Sounds like a really big deal.
Protegee's co-founders are Kirthi Banothu and Xiaoyu Li. Banothu already has a background in financial systems development and Li was previously a quantum trader at Optiver.
Vocera | Evaluations for AI voice agents
Vocera is the last but not the least startup on our list. This company offers services to those who are tired of manually testing advanced bots. Vocera automates the validation and scaling of voice agents for other businesses. Their AI simulates real-world scenarios using workflows, personas, and past conversations. As the founders say, the market is rapidly flooded with AI voice agents created by thousands of companies, and this startup is making them robust.
Vocera is spearheaded by Tarush Agarwal, Shashij Gupta, and Sidhant Kabra. They have experience in Google Research, quant finance, and working with Fortune 500 companies to make the new project a success.
These are the startups that I found most interesting. But YC has other AI-related projects as well. Here's a short list from F24 Batch if you want a stronger dive into the topic:
- Astrix Health - a toolkit for building safer health LLM applications
- Ember Copilot - The AI medical assistant
- Metreecs - an AI-powered demand forecasting for retail
- Manicode - a platform for code editing with AI
- Replexica - automated AI localization for multilingual software
Final Thoughts
So, AI agents. Which startup from our list do you think really deserves attention? Personally, I think that developers who are preparing the infrastructure for future adoption of digital assistants can play a really important role. And in this area, Vocera and Protegee seem to be the most significant projects from the new YC cohort.
Do you agree or do you have any other startups in mind? Tell us in the comments!
Share this edition with your friends!
This article was first published in the Creators AI newsletter. View the original edition.
