AIOps Startups funded by Y Combinator (YC) 2025

Fountain’s all-in-one high volume hiring platform empowers the world’s leading enterprises to find the right people through smart, fast, and seamless recruiting. Candidates can apply anytime, anywhere in minutes, right from their phone. Automated and customizable processes streamline the candidate experience and save time for recruitment teams so they can scale with growing hiring needs. Advanced analytics provide end-to-end process visibility so managers can make swift, data-driven decisions. Throughout the candidate journey, the openly integrated platform enables companies to find, qualify and convert more applicants. Fountain’s global customers hire over 1.2 million workers annually in 78 countries.

Prolific helps dedicated research teams to collect the very highest-quality human-powered data - at scale - using our simple-to-use platform to target and manage participants from our diverse, vetted participant pool. The truth matters: the best decisions, and biggest discoveries, are built on the highest-quality data. And with the increasing proliferation of AI, access to reliable, diverse data to develop and train AI models has never been more important. Created by researchers for researchers, Prolific was founded to provide a better way for researchers and organisations to get high-quality human data and feedback at scale for important research. Now, more than 30,000 researchers across academia and industry use Prolific to gather definitive human data and feedback from reliable, engaged and fairly-treated participants – with a new study launched every 3 minutes.

Rootly is a modern on-call and incident management platform used by top YC companies like Dropbox, Lattice, Webflow, Faire, but also Figma, LinkedIn, NVIDIA and 100s more.

Middleware is a full-stack cloud observability platform. We allow the dev and ops teams to debug the issue faster by bringing all the metrics, logs, traces, and events to a single unified timeline and train data to AI so it can give better insight into infrastructure and application. - Single command installation to collect all the metrics, logs, traces, and events. - AI-based anomaly and error detection - GPT 4-based error resolution - Support Infrastructure monitoring, Logs, APM, Kubernetes Monitoring, Synthetic Monitoring, RUM, Custom dashboard, and Alerts. - Full-stack observability, observe any stack at any scale. We are at Middleware and wanted to make cloud observability faster, cheaper, and better.

Sizeless is a platform to help ML teams automatically run, test and improve their models to make ML reproducible and safe. Currently, ML engineers do this manually which takes weeks - we automate this down to a few clicks. ML engineers at Google shopping and computer vision startups are using our platform to shorten their ML development cycle and continuously monitor and fine tune their models. We provide ML engineers with a productivity tool set, so that they can focus on building great ML models. We developed this tool during our PhDs (Cornelius in ML at ETH Zurich and Roger in astrophysics at Cambridge), saving us and our PhD colleagues around 2 weeks per model while generating actionable insights. We launched 10 days ago, have 5 daily active users like ML engineering teams at Google Shopping and in computer vision startups and more than 50 sign ups on our waitlist. We provide ML engineers with a productivity tool set, so that they can focus on building great ML models.

Logging is broken. Engineers waste hours in endless data. Sift Dev is building a future where you never read another log - starting with an AI pipeline to identify what matters and silence the noise.

Superb AI is a leading computer vision platform and professional services provider that provides enterprise-grade, end-to-end MLOps and DataOps workflows to accelerate the adoption and development of data-centric AI. Through the practical application of AI-based automation, Superb AI helps teams manage the entire ML lifecycle more efficiently, from data annotation to curation, model training, and deployment, while ensuring optimal data accuracy and consistency.

Automorphic has invented a way to infuse knowledge into LLMs via fine-tuning (surpassing context window limitations), enabling developers to rapidly iterate on and successively improve custom models cheaply and efficiently.

LLMs are incredibly powerful, but latency, cost, and unpredictable outputs have made productionizing LLM features challenging. Baserun is a testing and observability platform that helps AI teams streamline their development cycle from identifying an issue to evaluating their solution, so that teams ship faster with confidence.

Callback is AI-native business process outsourcing. We help our customers stay focused on their core business while we use AI to automate tasks like invoice processing, PDF report parsing, and data labeling.

1. Collaborative IDE for Building AI Agents 2. At its heart is a new programming language that merges plain English with concepts like loops, conditional statements, and function calling. 3. We also support forking, API deployment and sharing as a hosted app. We have 300,000 users, and many companies paying us between 199 and 10k per month.

Augento is the platform for DeepSeek R1 (like) fine tuning as a service. A user gives us an Agent and where it went wrong and we'll improve it. We provide a turn key solution to customers to act on poorly performing agents, by providing feedback in our UI to the agent and we'll take care of improving the model with reinforcement learning. Imagine a startup developing a tool for law firms, enhanced with an AI to assist lawyers in selecting clients. Currently, this startup would face significant challenges in building such a tool, as it’s crucial to provide accurate and consistent responses every time. With our technology, the startup can enhance the initial model by incorporating its own experience and feedback from beta users, eventually surpassing the initial model and delivering reliable results on every query.

As AI agents take on more complex tasks, ensuring reliability, accuracy, and efficiency is critical. Lucidic AI provides tools for AI agent devs, specifically visualizations and simulations, and engineers to test, debug, and optimize AI agents at scale, giving you full visibility into their decision-making. Modify prompts and re-run tests/simulations, watch step-by-step video replays of your agent’s execution, compare different models and configurations side by side, and visualize decision trees to prevent failures before they happen. Make sure your agents operate smoothly, adapt to changes, and perform as expected in real-world deployments.

We build magical data pipelines that use LLMs to integrate, clean, and enrich data living in disparate systems, delivering better KPI and financial reporting. We work with a wide range of customers, with a focus on PE- and VC-backed roll-ups where we provide a single source of truth for portfolio performance across acquisitions. Our experience is at the intersection of finance x tech. With an increasingly competitive M&A marketplace driven by $0.5T+ of dry powder, clean, usable data is a massive competitive advantage throughout the investment cycle – in diligence, to make informed investment decisions; in value creation, to execute the strategic plan quickly and confidently; and in exit, where better data leads to better outcomes. We’re on a mission to make AI useful for any company with messy data.

Tenyks (YC'S21) is a University of Cambridge spin-out inventing the intelligence that powers the AI revolution to protect the world from the Terminator. Imagine Snowflake for Vision, helping ML Engineers build better, safer models, in a faster and more efficient way. Specifically, our end-to-end platform is CV Engineers' Ultimate Productivity Tool to: 1. Automate Data Quality Detection: It streamlines the process of identifying and rectifying data quality issues, ensuring the integrity of your datasets. 2. Accelerate Data Curation: Our platform expedites the curation of similar images and objects, facilitating data collection and balancing. 3. Catalog Vast Datasets: With its flexibility, the platform allows you to efficiently index and catalogue extensive datasets. 4. Seamless CI/CD Integration: Tenyks effortlessly integrates with your CI/CD pipelines, reducing the risk of regressions and enhancing your development workflow. 5. Security & Scalability: Securely orchestrated in the cloud to foster seamless cross-team collaboration and scale to millions of images & hundred of millions of objects. The founders developed the technology (the result of 9 top-tier publications and a patented invention) during their PhD research in Safe & Explainable AI for Computer Vision.

Domu automates debt collection calls for the banks using generative AI. We have $870k in ARR with clients like BBVA, BNP Paribas and Skandia.

Mystic’s platform allows companies to easily and reliably deploy ML models on our serverless cloud or on their own infrastructure without requiring a team of MLOps engineers. With our Python SDK, data-scientists immediately get an endpoint from their own model, or any open-source models. Once uploaded, our platform handles routing, multi-cloud scaling, caching, GPU optimization and other features to provide the ultimate ML inference platform.

Our CLI makes ML model training effortless - just describe your job, and we handle GPU orchestration and execution at half the cost of major cloud providers.

Laminar is an open-source platform which provides observability, text analytics, evals and prompt chain management for AI agent.

PoplarML lets you deploy any machine learning model to a fleet of GPUs as a ready-to-use and scalable API endpoint with one command.

Athina is an end-to-end platform for product teams building production-grade AI features. 1. Prototype: Dynamic Columns let you run LLM prompts, execute code, make API calls, retrieve data, run transformations, and more (in a spreadsheet UI) 2. Evaluate: Run evaluations within the sheet itself - there are 50+ preset evaluation metrics or you can create your own custom evals (no-code). 3. Experiment: Change your prompt, model, retriever, or chain in just a few clicks. Technical users can run even more experiments programmatically. 4. Compare: Compare any number of datasets side-by-side with a diff view, and run evaluations on all of them simultaneously 5. Observability: Monitor your LLM features in production with continuous evaluations and granular analytics. Athina can be used by non-technical users as well as technical users. This means PMs, QA, Data Scientists, and Engineers can collaborate to build great AI products.

Pipeshift is a modular orchestration platform for building with open-source AI components — embeddings, vector databases, LLMs, vision models and audio models — across any cloud or on-prem. Enterprises get to deploy their AI workloads in production faster and more reliably. Additionally, as we see more model and hardware architectures coming in the market, Pipeshift future-proofs the infrastructure investments by offering flexibility through their modular MLOps stack that allows enterprises to bring down their GPU infrastructure costs without any additional engineering efforts on their end.

Helicone.ai is creating an advanced observability platform tailored for developers working with Large Language Models (LLMs). Our goal is to simplify and enhance the operational side of deploying these models, making it easier for developers to monitor, manage, and optimize their AI applications at scale. Helicone provides a unified view of performance, cost, and user interaction metrics for various LLM providers, like OpenAI, Anthropic, and LangChain, empowering developers to make their LLM deployments more efficient, reliable, and cost-effective. ### Key Features 1. **Centralized Observability**: Our platform captures and visualizes detailed logs and metrics across all LLM deployments. With tools for prompt management, performance tracing, and debugging, Helicone provides real-time insights into the inner workings of your LLMs. 2. **LLM Performance Optimization**: Helicone supports prompt experimentation, success rate tracking, and fine-tuning, allowing you to continuously improve response quality and efficiency. This level of insight makes it easier to deliver high-performing, cost-effective AI applications. 3. **Flexible Data Management**: We understand that data privacy is critical. Helicone supports deployment options for dedicated instances, hybrid cloud integrations, or self-hosted environments, allowing clients to maintain control over their data and ensuring compliance with privacy standards. ### Built for Developers and Data Scientists Helicone is designed to meet the needs of engineers and data scientists who require transparency and control over their LLMs. From chatbots to document processing systems, Helicone equips you with the insights needed to track costs, understand user interactions, and optimize outputs—all from one intuitive platform. By combining observability with LLM-specific insights, Helicone is redefining AI monitoring, empowering developers to deploy and scale their AI models with confidence.

Vespper is an AI agent running 24/7 to troubleshoot your alerts by surfacing the right data at the right time. It helps teams reduce downtime and alert fatigue by making sure no alert falls through the cracks again.

Abundant powers reliable and safe AI deployment via our API. Our network of specialized operators works alongside AI agents to: (1) Handle complex edge cases with precision. (2) Maintain rigorous safety standards. (3) Generate high-quality training data that continuously improves AI performance. Companies can meet stringent quality requirements at a fraction of the cost of in-house teams, while accelerating their AI initiatives.

Outerport helps AI read documents the way a researcher would - building up understanding section by section and checking facts along the way, instead of just matching similar phrases. For example, when a team needs to confirm a specific requirement in a 200-page compliance policy document, Outerport pinpoints the right section, provides the reasoning, and presents a fact-based summary. It can leverage this deep understanding to perform more complex tasks over documents, like comparative gap analysis and change tracking. Outerport is built for performance- run thousands of queries against your critical documents, with reasoning behind every retrieval. No more faulty vector RAG.

Ryvn manages the infrastructure for your applications in your cloud or your customer's cloud. We handle deploying and rolling out updates, provide aggregated logging and monitoring, and offer one-click rollbacks. Expand into regulated industries, accelerate time-to-PoCs, and boost revenue by offering on-prem/self-hosted without the added complexity. Build once, deploy anywhere.

We build tools that allows enterprises verify information about a user *before* processing it - whether it is processed by a human or an AI agent. Traditionally, these verifications including but not limited to background checks cost ~$10. Using our advancements in AI and Cryptography, we're able to send that cost to a few cents. This enables enterprises to do verification at scale, earlier in their work-flows, and enables new kinds of businesses - which were not performing these verifications because of prohibitively high costs. For example, it makes no sense that a recruiter does a candidate's document verification and background checks AFTER the candidate has cleared all the interviews. That's because that verification is cumbersome and expensive. They can't afford to verify all the candidates at the top of the funnel. Many such workflows are ripe for disruption. Another example of a product that was not possible before. A dating app uses our tools to verify the users' information - including their employment, education and financial background. This was something that would have been too expensive to verify at a $10 pricepoint where the ARPU is only $8. Not only that, we're designing the tools ground up - so that AI agents can utilize these verifications before ingesting information, so that they take action basis truthful information. We have 3 tools 1. BlueCheck : our flagship product for background verification. The cheapest option to verify users' employment, education and financial background. A few cents instead of a few tens of dollars. Several Billion Dollar+ revenue companies are customers. A solution designed for enterprises that works out of the box! 2. Reclaim Protocol : our underlying infrastructure that makes BlueCheck possible. With advancements in cryptography, specifically zero knowledge proofs - our research over the last 3 years have resulted in inventing a new cryptographic primitive called zkTLS, which allows us to verify information from any website using the cryptography of the existing internet pipes (TLS). This is our secret sauce! 3. Questbook : our dog-fooding tool. Questbook is a tool to give out research and open source software development grants. These grants are reviewed by AI agents that we built inhouse. The biggest scam people try is to fake who they are when they are applying for grants. We dog-food our own tools to make sure we verify all the claims the users are making when they apply for grants. Our tools secure more than $10M of grants being given out on the platform. So it's high stakes, and it works!

Maitai makes building reliable AI applications easy. We autocorrect faulty model output in real-time and automatically fine-tune models that learn from their mistakes. This means our customers get more reliable results immediately, and over time, they gain custom models built specifically for their application that only get better and faster. You wouldn’t hire an employee who doesn’t learn from their mistakes—so why use a model that doesn’t? Maitai is here to deliver the next generation of reliable AI inference.

OpenPipe is an SDK that abstracts away fine-tuning custom models. We capture your existing provider’s prompt-completion pairs in the background and use them to create a new model that is faster, cheaper and often more accurate than the original.

🚀 Lumina💧 (YC W22) is a fast growing job platform in Southeast Asia with 1.5 million workers, helping 30k companies recruit and optimize at scale. 🤖 We are now expanding globally with Lumibot, a bespoke AI+human-powered automation service. We help companies automate mundane, repetitive processes with AI and human-in-the-loop catering for various use cases in data ops and complex workflows. This allows your business to focus on your core and grow faster, better and cheaper. 👌Our AI-powered automation start with USD 13/hour, supplemented with human-in-the-loop at USD 2-6/hour. Our services are managed based on SLAs by dedicated service managers under a prepaid credit system without arduous contracts for optimum flexibility. No minimum commitment with fully-refundable deposit from as low as USD 500^. 🛠️ We currently support the following use cases: 1) operations support 2) revenue operations support 3) web and social media management 4) email management 5) content moderation 6) AI training and verification 7) AI text humanization 8) custom process automation ... with more to come ... 👥️️ Lumibot (by Lumina) is backed by Y-Combinator, Monk's Hill Ventures, Alpha JWC, Goodwater Capital, SWC Global and January Capital. More information about us: * 💬 WA: https://wa.me/message/SEMGBRM22MDYK1 * 📅 CAL: https://usemotion.com/meet/aswin-andrison/global * 🌐 WEB: https://www.lumibot.com ^ Reach out to us on special deals ONLY for YC companies!

RedBrick AI's mission is to accelerate the adoption of artificial intelligence in radiology by building world-class software infrastructure. Currently, we are focused on helping radiology AI teams prepare high-quality datasets to train their algorithms. Radiology data, such as CT and X-ray scans, is an incredibly important source of truth in healthcare delivery. In fact, over 90% of all healthcare data is medical imagery! However, the global radiology workforce is overburdened. In the UK, for example, only 2% of radiology departments are able to fulfill their reporting requirements, and this trend is reflected worldwide. The acute state of radiology, coupled with the abundance of data, makes the use of AI in radiology a prime candidate. In 2022, $5.6 billion was invested in the development of AI in healthcare! However, a key hindrance to further adoption is the lack of sophisticated tools to build and deploy AI algorithms in clinical environments. This is the problem we’re focused on at RedBrick AI We're a team based out of Bangalore India, and USA. We're backed by leading institutional investors like Y Combinator and Peak XV (formerly Sequoia Capital India).

LiteLLM is an open-source LLM Gateway with 18K+ stars on GitHub and trusted by companies like Rocket Money, Samsara, Lemonade, and Adobe. LiteLLM provides an open source Python SDK and Python FastAPI Server that allows calling 100+ LLM APIs (Bedrock, Azure, OpenAI, VertexAI, Cohere, Anthropic) in the OpenAI format We have raised a $1.6M seed round from Y Combinator, Gravity Fund and Pioneer Fund

Openlayer is the governance platform for AI applications.

Keywords AI is used by 40+ fastest growing YC startups to ship reliable AI features really fast. Monitor and debug: - Request logging with full text search to monitor AI in prod - LLM agent tracing to debug workflows with ease - Metrics dashboard to understand usage, cost, latency, and get product growth metrics - User analytics to learn how every user is using your product Build and iterate: - Collaborative prompt management - Playground for 350+ models - Model gateway with OpenAI/Anthropic API interface - Eval & experiments

LanceDB is a new open-source vector database that can support low-latency billion-scale vector search on a single node. Built around a new columnar data format, LanceDB makes it incredibly easy to build applications for generative AI, recsys, search engines, content moderation, and more.

SubscriptionFlow is an AI-based customizable revenue management platform that helps SMBs optimize subscription revenue. It monitors user activity to identify customers at risk of churning and automatically alerts the sales team. For example, eLearning/SaaS/Membership businesses require to monitor subscription usage to improve customer retention. Ecommerce/Magazine companies need to see the subscription trends to cross-sell and upsell products/services.

Query Vary is a no-code tool for building LLM-powered automation. Example use cases include extracting information from a sales call transcript and piping it into a CRM, creating a Slack bot trained on product documentation, or transforming Google forms into fully formatted reports. Input Software --> LLM --> Output Software Our extensive toolset to make each workflow reliable sets Query Vary apart. Book a demo: https://calendar.app.google/6oxEvZA2k4kK6dBa8

Airtrain AI is a no-code data platform for Large Language Models. Proprietary AI models such as GPT-4 are very powerful but also very costly, slow, unreliable, and unsecured. As businesses look to scale their AI prototypes into production-grade products, they struggle with large AI bills, slow APIs and large failure rates. On the other hand, smaller language models have been proven to be able to perform on-part with large ones with fine-tuned on high-quality datasets. Airtrain AI lets AI practitioners explore alternatives to proprietary models, build up training datasets, evaluate, fine-tune, and serve a large selection of open-source LLMs.

NuMind is a tool for data scientists, data analysts, but also software engineers to create custom NLP models. For example, a recruiting company uses NuMind to find which job offers best match a given resume. Etienne (CEO) was head of Machine Learning at Wolfram Research, and Samuel (CTO) co-founded Make.org (8M users). NuMind originated from our own frustration when developing NLP models. Leveraging large language models similar to GPT-3, NuMind allows to complete NLP projects at least 10x faster than before. We launched a private beta August 2 and had 9 paying customers one month later.

Chatter is Postman for LLMs. Our platform helps companies and developers test their LLM models. Iterate on prompts, run them across model families and evaluate them against test cases – all in one place. With collaboration features, engineers can design LLM chains while QA can write test cases.

We’re on a mission to enable everyone to build reliable AI agents & AI applications. We're fully open source, and provide a unique integrated development & observability experience for those building anything in the AI space. This is the first step towards laying the foundations for Composable AI Systems; all AI systems need observability and introspection to be first class for them to be reliable. How? We're standardizing how people write python to express data, ML, LLM, & agent workflows/pipelines/applications with lightweight frameworks. So that no matter the author, it'll be easy to collaborate, connect, and importantly in one line integrate observability and datastore needs. This speeds up time to production and reduces TCO because code remains easy to maintain and your data flywheel stays manageable. So you can increase the top line & bottom line of your business by delivering on AI that is reliable. We've got two open source projects: - one focused on AI applications, called Burr (https://github.com/dagworks-inc/burr). - one focused on AI pipelines/workflows, called Hamilton (https://github.com/dagworks-inc/hamilton) see https://www.tryhamilton.dev Both Hamilton & Burr come with self-hostable UIs (+ enterprise & SaaS offerings). With a one-line code change, you get versioning, lineage / tracing, cataloging, and observability out of the box with Hamilton. With Burr you get tracing, observability and persistence in a single line addition.