Infrastructure Startups funded by Y Combinator (YC) 2025

Fivetran automates data movement out of, into and across cloud data platforms. We automate the most time-consuming parts of the ELT process from extracts to schema drift handling to transformations, so data engineers can focus on higher-impact projects with total pipeline peace of mind. With 99.9% uptime and self-healing pipelines, Fivetran enables hundreds of leading brands across the globe, including Autodesk, Conagra Brands, JetBlue, Lionsgate, Morgan Stanley, and Ziff Davis, to accelerate data-driven decisions and drive business growth. Fivetran is headquartered in Oakland, California, with offices around the world. 

CoreOS is part of the Red Hat family. A leader in the open source community and an innovator of open source projects, CoreOS creates and delivers critical components for containerized applications that help fuel broad adoption of secure, scalable and resilient infrastructure. The team maintains several open source projects, including: Container Linux: the secure automatic updating OS; and etcd, the distributed data store for Kubernetes. Join us at https://coreos.com/ and @CoreOS on Twitter.

Heroku, a Salesforce company and industry pioneer in platform as a service (PaaS), enables developers to build and run applications entirely in the cloud. Heroku provides companies from startups to Fortune 500 enterprises with a faster and more effective way to create, deploy, and manage apps. Heroku operates the world’s largest PaaS, continuously delivering millions of apps with 6+ million container deployments, 60+ billion routing requests, and 10+ terabytes of application logs per day.

Texel is a platform that enables developers to generate hardware-accelerated APIs for running AI pipelines 10 times faster, including image and video applications. For example, developers can use the latest NVIDIA GPUs or Google TPUs without having to become hardware experts themselves.

SuperAPI is a programmable API gateway that makes all your fetch APIs cacheable, reducing the response time 10x. Our customers use SuperAPI to cache their critical APIs without spending 6-9 months rewriting their code.

Expo lets developers write apps that are written in JavaScript and run natively on Android, iOS, and the web. Our cloud services that work with all Expo and React Native apps, named Expo Application Services (EAS), further speeds up teams by helping them get to the app stores as fast as possible and continue growing. We believe that mobile apps provide a better experience than websites, but making apps is much more difficult than it needs to be. With Expo, we're bringing together the best of the web, like its development speed, fast iteration cycles, easy deploys, and automatic cross-platform support, and the performance, device capabilities, and user experience of native apps.

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 are working on the software that will power the next generation of Cloud Computing (Edge Computing) platforms using WebAssembly. You can see us as a mix between Docker (but 100x more performant in startup time, and with 100x more lightweight containers), npm (we released the first WebAssembly package manager: wapm.io) and the JVM (but more universal).

We build software that helps hardware teams quickly and clearly evaluate how a component (engine, battery, valve, etc.) or process (assembly line, manufacturing plant) performs. We prevent mistakes, reduce costs, and accelerate time-to-market. Our platform unifies industrial control with sensor data storage and automated post-processing, delivering an end-to-end operations and observability stack. To put it comparatively, take LabVIEW, InfluxDB, and Grafana, wrap them into a single package, add a bit of magic, and you get Synnax.

We’re building the an end-to-end web simulator for browser agent training and evals. Our system enables teams to test, benchmark, and optimize browser automation models at scale. - Deterministic Web Simulation → Stable, reproducible testing with versioned web snapshots. - Live Web Evaluation → Identify failures caused by UI drift, captchas, and dynamic content. - Automated Annotation & Labeling → Generate high-quality training data for benchmarking. - RL-Driven Agent Optimization → Improve models with scalable, feedback-driven learning. By combining synthetic user simulations, automated evaluations, and large-scale benchmarking, we help teams build more reliable web agents that handle real-world environments with confidence.

Orchestrate custom integrations, workflows & agents in hours, not months. Take control of your integration strategy and breathe easy with maintenance on AI Autopilot. For Product Teams: Build better integration libraries. Build a feature-rich integration library, for your users to enjoy. Offer out-of-the-box integrations that work for you and your customers. Embedded IPaaS typically locks you into connector or endpoint limitations. Versori gives you to tools for limitless customisation. Proactive, self healing agents, scan your connected apps for endpoint or schema changes. You get alerted, Versori AI fixes the change. Embed Versori built integrations into your app with the Versori SDK. Flexible to your development approach with advanced user management. For Operations Teams: Get your internal systems speaking the same language. Deliver integrations for new software in days, not months—so you can start unlocking value immediately. Versori’s speed to value reduces typical deployment fees by half—or more. Low code for speed. Full code for control. No more limits from inflexible integration platforms. For GTM & Sales Teams: Say yes to any prospect's integration request. Stop bouncing between teams to get integrations built. With Versori, Sales can go straight to yes. No more escalations or delays. Versori offer fully managed custom-builds, so your customers get exactly what they need, without compromise.

14.ai is a Zendesk alternative built from the ground up for AI agents. The Cursor for customer support.

We are making it possible for digital businesses to easily verify individuals and businesses within seconds across Emerging Markets.

The fragmented and proprietary evaluation tools today are leading to significant inefficiencies and confusion among developers. The world needs a standard everyone can rely on and that is why we are building Ragas as the open-source standard. We have 4k stars on GitHub, 1.3k members in our discord community, and over 80+ external contributors. We also have partnerships with key AI companies like Langchain, Llamaindex, Arize, Weaviate and more to help create a standard. We already process 5 million evaluations monthly for engineers from companies like AWS, Microsoft, Databricks, and Moody’s and it is growing at 70% month over month. We are building LLM application testing and evaluation infrastructure for Enterprises.

AssemblyAI is an applied AI company. We research, train, and deploy State-of-the-Art AI models for developers and product teams to integrate into their products/apps. Today, we offer AI models for audio transcription, content moderation, summarization, topic detection, and many other tasks. We are working towards building the "Stripe for AI models" - where developers and product teams will be able to easily access State-of-the-Art AI through a simple API. We're a fully remote team of AI researchers and ML engineers, and have raised over $65M in funding from top investors including Accel, Insight Partners, Y Combinator, John/Patrick Collison, Daniel Gross, and Nat Friedman.

Web3 developer platform to build, test, and scale blockchain applications. Lightning fast access to 44+ chains like Ethereum, Bitcoin, Solana, Polygon, Optimism, Arbitrum, Avalanche, Near, Stacks, Flow, Tron, zkSync and more. Elastic APIs, advanced ETL, storage (IPFS), powerful tools, analytics, and a 3rd-party marketplace -- all at your command through a simple control panel. Build your next big thing on-chain! Build it with QuickNode.

Provision is building the contract analysis platform for construction. Instead of manually reading through thousands of pages of documents and revisions, Provision organizes and extracts information so constructors can save time and reduce the cost of future mistakes.

Aden is an AI-native, multi-agent ERP platform built for operations leaders to unify their business systems, streamline processes, and efficiently manage objectives. Highly customizable and rapidly deployable, Aden boosts Key Performance Indicators (KPIs) with declarative solutions and intelligent AI agents. Currently, Aden helps over 100 operationally-focused businesses, including supply chain control towers of major global corporations, IT departments of major financial institutions, and other high-impact companies, elevate their efficiency, automate workflows, and scale beyond manual operations.

Coris is a Risk API for SaaS companies and Fintechs that underwrite SMBs. We built the risk systems at eBay, PayPal, WePay, and Google. Leveraging our expertise, we were able to build the product in less than 3 months, launch it within few weeks, and already have live pilots with companies such as Housecall Pro and Wave Accounting

Inscribe’s award-winning Risk Intelligence technology is the most effective way for fintechs and banks to automate manual reviews. Inscribe can reduce fraud losses and operational costs while increasing the accuracy and efficiency of risk decisions. Founded in 2017, Inscribe is trusted by leading fintechs like TripActions, Plaid, Bluevine, and Ramp. Inscribe is backed by Threshold Ventures, Uncork Capital, Y Combinator, Crosslink Capital, and Foundry, and is headquartered in San Francisco, CA. Learn more at inscribe.ai.

Downlink is a platform that continuously improves your AI. Boost rate limits, slash latency, and raise accuracy.All through one simple API.

At Aptible, we envision a world where developers achieve more with less infrastructure. We’re building the platform that empowers developers at companies of all stages of development to move beyond the traditional method of managing infrastructure—DIY with expert infrastructure and platform engineers on AWS, GCP, or Azure. We’ve done this through our secure, compliant, reliable, and scalable Platform as a Service that’s been used by thousand of developers, hundreds of companies, and dozens of publicly traded and unicorn companies since 2013.

Our customers use Freshpaint to make their websites and entire marketing stack HIPAA compliant. It's impossible to use modern analytics and marketing tools without sharing sensitive HIPAA-regulated data. Our platform does the hard thing of moving data from A to B. And then we do the really hard thing of controlling the flow of sensitive HIPAA-regulated data to various 3rd-party marketing tools. We're building the data infrastructure that safeguards patient privacy while also enabling marketing teams promote access to healthcare.

Infrastructure for search teams building retrieval and RAG. Trieve combines search language models with tools for tuning ranking and relevance. Building excellent search is difficult and can take months to implement then even more time to maintain. Trieve offers production-ready infrastructure that works out of the box to help search teams build adjustable AI search and RAG into their products. With tools for custom models, relevancy weighting, date-recency biasing, semantic full-text hybrid search, recommendations, and more Trieve covers the full spectrum of what a team wants from their discovery infrastructure.

Blacksmith helps companies run their continuous integration (CI) up to twice as fast, at half the cost. We do this by running their CI on high performance gaming CPUs. - Engineering productivity is often marred by slow CI that grows quadratically with the size of the engineering organization. - On top of this, companies pay high markups to run CI on the hyperscalers. - Consequently, companies have to run their CI on spot instances to keep costs in check. Since spot instances can get pre-empted at any point, running CI reliably on them requires [significant engineering investment](https://www.rippling.com/blog/how-rippling-used-spot-instances-to-save-and-scale-ci-cd). Blacksmith offers CI compute that is more performant, cheaper, and reliable. With Blacksmith, companies do not have to think about CI as they scale. We are starting with GitHub Actions - today's most popular and fastest-growing CI system. With a one-line change, companies can run their CI up to twice as fast on our high-performance hardware. In addition to this, Blacksmith provides colocated warm caches along with optimizations for common CI workloads (like building and pushing Docker images). Aayush, JP and Aditya have gained deep systems and product knowledge from their time at Cockroach Labs and Faire respectively. Both these companies, with several hundred engineers, were spending in the order of millions of dollars on their CI. Yet they were riddled with slow CI and actively seeking out cheaper and faster alternatives. Blacksmith offers exactly this alternative, at a time where being lean is table-stakes for most companies.

Increase engagement, conversion, and revenue with a configurable search and recommendation system that adapts in real-time.

RunRL improves LLMs and AI agents with reinforcement learning. Do you have a model that needs to get better at a certain task? Are you tired of messing around with prompts? Do you spend a lot of money on observability, and wish that all this data could let your model self-improve? Well, head on over to RunRL.com, and see what we have to offer! If you give us a model, a prompt, and a reward, we’ll make your model’s reward go up.

Pylon provides water and electricity distribution companies, in emerging markets, with a platform for reducing their losses as well as opening new revenue streams. These losses are coming from leakages, consumption thefts, and uncollected bills, amongst others. We have signed 11 utilities onto our platform across 2 countries and connected our systems to over 560k endpoints. Today our average client has reduced their aggregate revenue losses by 45% Expanding in Africa, LATAM, and SEA.

Mezmo, formerly LogDNA, is an observability platform to manage and take action on your data. It ingests, processes, and routes log data to fuel enterprise-level application development and delivery, security, and compliance use cases. Mezmo was brought to life by three-time co-founders Chris Nguyen and Lee Liu and included in the Winter 2015 batch of Y Combinator. In 2018 the company partnered with tech giant, IBM, to become the sole logging provider for IBM Cloud. Mezmo is on a mission to empower people who build solutions that shape the world. We’re doing this by delivering a platform that enables enterprises to get more value from their observability data in real time, regardless of source, destination, use case, or scale. We’re not the only ones working on this problem but we have a few things the others don’t. We’re cloud-native and know how to make the most of modern technology like Kubernetes. We have scaled a solution from zero to petabyte scale in a short amount of time, while supporting thousands of active users across multiple environments. We are hungry for change and are surrounded by enterprises telling us they’re hungry, too. We have a kick-ass group of people who are thinking about the problem analytically and are excited to change the observability world for the better. Mezmo has helped some of the world’s most innovative companies transform how they manage their systems and applications. Still, we know that we can help them get more value from their observability data by providing more flexibility and control over how they use it. This will enable teams to spend less time switching between data silos so they can focus on shipping better, more resilient, and secure products. We have momentum on our side. Last year we saw triple digit revenue growth and added 800 new customers to our roster. Recent accolades include being named to YC’s Top Companies, CRN’s 10 Hottest DevOps Startups, and EMA’s Top 3 Observability Platforms.

The IOMETE self-hosted data lakehouse platform unifies all data - regardless of where the data resides - enabling large-scale analytics (BI/ML/AI) on the entire data set. IOMETE runs always in the customer's environment, whether on premises, on public cloud or private cloud thereby maximizing data security, data privacy and data ownership while reducing cost.

Hiro System builds developer tools for Bitcoin L1 and the Stacks L2 (layer-2). Stacks is a Bitcoin layer for smart contracts; it's an open-source project with an ecosystem of independent entities. Hiro serves 350M+ monthly API requests and thousands of crypto users on dev products.

Sunpia lets developers easily experience the cost and speed benefits of serverless infrastructure, without having to rewrite their code. Developers annotate their code and Sunpia automatically designs a microservice version of it they can deploy on their own cloud.

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 AI agent helps teams save money and ship faster on AWS, GCP, and Azure. Visit infra.new to try free without signing up! - Generate reliable cloud configuration - View detailed cost estimates and system diagrams - Optimize everything with AI

Virtualmin is taking on the lack of innovation in the server admin programs, like Plesk, by making a more accessable version for pages managed by the non-technical crowd. The program will feature simple installs of popular programs like content management systems that often cost extra on other providers. It will also let you administer your website from your desktop and mobile device.

Copilot's mission is to power the next generation of modern service companies. We help marketing agencies, accounting firms, and many other types of businesses to sell and deliver their services online.

Open-source alternative to FullStory and LogRocket. Self-hosted for a complete control over your data and costs.

Wordware is an IDE for building AI agents using natural language. It looks and feels like Notion, but lets you design, test, and deploy AI systems in real time - without writing code. Our mission is to bring structure and joy to human–AI collaboration. We’re building a generational company that empowers the next billion knowledge workers to create with AI - not by writing code, but by expressing intent. We’re backed by Spark Capital, Felicis, and Y Combinator ($30M seed round - the largest in YC history). We work hard, move fast, and don’t take ourselves too seriously. It’s intense, but it’s also fun - at Wordware, you’ll do the best work of your life alongside people you genuinely like.

Pluralith automates compliance by documenting cloud infrastructure directly from the codebase. Companies like Jefferson Health pay us to automate their infrastructure compliance and save 5h a week per infrastructure engineer. More simply put: We generate infrastructure diagrams directly from Terraform state.

Eventual is a data warehouse for ML/AI. We are building an integrated development experience for data scientists and engineers to build ML/AI applications which query and process multimodal data (images, video, audio and 3D scans). Daft (https://www.getdaft.io) is our open-sourced distributed data query engine. It has a Python Dataframe API and is built in Rust, providing both ease of use and performance for use-cases such as ML data engineering, training data ingestion and general data analytics. We are funded by investors such as Caffeinated Capital, Array.vc and top angels in the valley from Databricks, Meta and Lyft. Our team has deep expertise in high performance computing, big data technologies, cloud infrastructure and machine learning.

Video has become a core component to some of the most consequential industries in the world including social, communications, art, robotics, and VR. But the processing of it, especially with AI, is incredibly complex due to the number of use cases that exist and the types of systems that need to be built for each one of them. Sieve's mission is to supply the infrastructure needed to understand, manipulate, and generate video, so we can accelerate all future applications of it.

Netmaker is an open source virtual networking platform based on WireGuard. With Netmaker, you can bridge, clouds, the edge, IoT devices, and Kubernetes clusters. Netmaker provides the fastest possible virtual networks, with an average speed 10x faster than standard alternatives. Think of Netmaker like your VPC for distributed systems.

Protocol Labs is an open-source research, development, and deployment laboratory. Our projects include IPFS, Filecoin, libp2p, and many more. We aim to make human existence orders of magnitude better through technology. Some of our projects. IPFS makes the web work peer-to-peer. IPFS is a protocol for storing, addressing, and transferring data, designed to replace HTTP in many contexts. IPFS uses content-addressing to uniquely identify and link to information. IPFS is one of the leading protocols of Web3, crypto, and the decentralized web movements. IPFS serves millions of end users and tens of thousands of developers. It powers the websites and applications of small startups, large enterprises, and even blockchains. Filecoin is a decentralized storage network, designed to store humanity's most important information. Filecoin is a cryptocurrency incentivized decentralized market, where storage providers earn filecoin tokens for renting out storage space. (Think airbnb for cloud storage). Filecoin is one of the leading projects in the cryptocurrency/blockchain space.

Decent's core product,[The Box](https://decent.xyz/), enables 1-click transactions using any token across chains. For example, a user can purchase an Optimism NFT off primary or secondary sales using USDC on Arbitrum. As applications splinter across an increasing number of rollups and app chains, wallets and dapps will have to meet users at their point of liquidity vs. expecting users to jump through hoops to arrive with tokens on the correct network.

Docker is the world’s leading software container platform. Developers use Docker to eliminate “works on my machine” problems when collaborating on code with co-workers. Operators use Docker to run and manage apps side-by-side in isolated containers to get better compute density. Enterprises use Docker to build agile software delivery pipelines to ship new features faster, more securely and with confidence for both Linux and Windows Server apps.

RDash is a new age construction management platform for Developers, contractors and design studios. It enables complete workflow automation for design, projects, procurement, and finance professionals involved in construction and interior projects management.

Neosync is an open-source platform that allows you to create anonymized or synthetic data and sync it across all environments for testing and machine learning. Companies in highly regulated industries such as FinTech, HealthTech, InsureTech and those with sensitive data can use Neosync to create production-like data to use for debugging and building features in lower-level environments without the security and privacy risk of using production data.

Porter is the easiest way to deploy and scale your applications on AWS, Azure, and GCP. Connect your own cloud account, and Porter will provision and manage the underlying infrastructure for you - simply point at your repository and Porter will handle the rest, from building your application to autoscaling it. Under the hood, Porter manages the same kind of infrastructure that growth-stage companies use and allows you to scale without DevOps overhead.

AccessOwl automates employee SaaS account provisioning and permissioning for any tool. IT teams from companies like Zeplin and Sary save 30 minutes on every request. Access management used to be an enterprise problem. With work from home, exploding SaaS usage and the need for SOC2 certifications startups have the same problems as enterprises. Mathias and Philip have built several startups and teams before and experienced the same problems around access management over and over again. Philip was the CEO and Co-Founder of blik, an award winning IoT solution that was used by Volkswagen and BMW. Mathias built the leading private payment infrastructure provider in Germany as well as the first digital car insurance as Co-Founder and CTO. Based on these experiences we decided to build a solution we wish we could have used when building our former companies. AccessOwl can be deployed within minutes instead of weeks by simply installing a Slack application. We save companies that use AccessOwl 30 minutes for every access request, on- and offboarding.

Plutto gathers all the information and documents needed to verify and identify any business in LatAm. We process this information and provide insights to your on-boarding or compliance teams so they can make better decisions. This increases on-boarding speed from weeks to minutes and automates highly manual processes such as requesting and reading documents.

LLMs run riot in production. Get back in the driving seat. Build your own evals, iterate quickly, and go from prototype to production in no time! Try Now! https://console.unify.ai/ ✨

​​Cortex is an internal developer portal built to accelerate the path to engineering excellence. Companies like Xero, Canva, and TripAdvisor use Cortex to catalog, score, and assign action to improve service quality and velocity, so devs can get back to work that drives the business forward.

What if you could build web apps as easily as you make a Figma design or edit a Google Doc? Enter Mocha. We've reimagined app development with a state-of-the-art coding agent that builds for our users. Want to launch that app idea? Need to build something for your community? Looking to prototype faster? Start here. We've removed the barriers between idea and execution. Our mission is to empower everyone to build for the web, our vision is personal software.

Building the best AI agent co-worker by giving agents access to tools & the internet

Odigos leverages eBPF technology to create an open-source solution that can automatically generate distributed traces for any application without any engineering work. Developers are struggling with extracting and delivering high-quality monitoring data. Currently, building a pipeline is a manual process that takes months to implement, is an error-prone process that requires tons of code changes across multiple R&D teams, and demands collaboration.

Start your AI application with agentic RAG, ingestion, document management, and more.

Janus tests conversational AI Agents for hallucinations, rule violations, tool call failures, and performance breakdowns pre-launch. We do this by simulating thousands of realistic AI users to eliminate manual testing and low-confidence in AI deployment. Our platform generates personalized datasets to evaluate, benchmark, and improve models over time.

Morphik is open-source multimodal search for AI apps. We're used by customers ranging from space-tech teams searching across research papers to developers building brokerage agents.

# Our Mission 🎯 At Shuttle, we’re on a mission to revolutionise developer experience for building web apps. Everything we do is about crafting an exceptionally simple, satisfying and effective way to do cloud development. We are open source community-driven and loved by thousands of engineers. Our vision: the ultimate platform to supercharge developers’ productivity. # About Shuttle 🚀 We are building the fastest backend development experience. [Shuttle](https://shuttle.rs/) is an open source cloud platform for developing and deploying apps with zero infra hassle. At [Shuttle](https://www.shuttle.rs/), we're on a mission to redefine the developer experience for building cloud applications. We want to empower the next generation of builders to make use of the latest and greatest cloud technologies, by crafting the simplest, fastest, and most enjoyable development experience. Shuttle is open source and our [Infrastructure from Code](https://www.shuttle.rs/blog/2022/05/09/ifc) framework paired with our cloud platform eliminates the infra hassle and accelerates app development and deployment. This is a game-changer for developer productivity - think Vercel for the backend. We have thousands of developers in our community incl. engineers from some of the top companies and have just launched Shuttle beta. We’re backed by some of the best investors in the world, including Y Combinator, HorizonVC, Abstraction Capital, Global Founders Capital, Founders of Segment, Sendbird, Head of Product at DataDog and other successful founders and executives. # Working With Us 😎 We’re a growth mindset company. We’re honest, open-minded, driven, and ambitious and we are only looking for people who share these qualities with us. We care. About each other and about our mission. We are building a culture that brings out the best in people and we truly believe that the company you work at should be the launch pad for unleashing your full potential. We keep a light and fun work environment, without unnecessary stress or anxiety. Yet we hold ourselves to very high standards and keep pushing the bar up. Continuous improvement is a must if you don’t want to be left behind. Mistakes are OK, not being willing to learn from them is not OK. Collaboration is at the core of what we do and we treat each other with respect and kindness - supportive, yet firm and ready to provide accountability where needed. We’re early stage so expect many opportunities to shape the company’s future. The best ideas win. We maintain a hacker mentality - no problem is unsolvable with the right mindset. As a team, we work on hard challenges under tight constraints and are constantly pushed out of our comfort zones - if reading this excites you, let’s talk. # Perks 🎁 - Remote-friendly, flexible working hours and workdays - Flexible holidays and paid time off - Mental and physical health stipend - Educational and self-development budget - Support for attending and presenting at relevant conferences - Health insurance - Awesome work-from-home setup & equipment - Regular company off-sites - Regular remote-friendly socials (games and more)

SigNoz is an open-source observability platform native to OpenTelemetry with logs, traces and metrics in a single application. An open-source alternative to DataDog, NewRelic, etc. Open source Application Performance Monitoring (APM) & Observability too

Svix is the enterprise ready and open source webhooks service. Webhooks are a pain. Developers need to worry about deliverability, retries, monitoring and security. All of which are different for webhooks compared to the rest of the stack. We turn all of that into a simple API call. Svix is sending billions of webhooks for customers both large (Fortune 500) and small (startups), and is backed by Y Combinator, Andreessen Horowitz (a16z), Aleph, and founders and CTOs of companies such as Github, PagerDuty, Segment and Lookout.

Thunder Compute is the world's cheapest GPU cloud platform for researchers, data scientists, and startups. The team is building a proprietary orchestration stack to improve GPU utilization by 5x. Try it today.

We provide a simple API for creating, storing, versioning, and collaborating on multi-modal AI datasets of any size. With Activeloop's open-core stack, you can rapidly transform and stream data while training models at scale. Deep Lake powers foundational model training by acting as a vector database with significant benefits, such as (1) the ability to use multi-modal datasets to fine-tune your own LLM models, (2) storing both the embeddings and the original data with automatic version control, so no embedding re-computation is needed (3) truly serverless service with no vendor lock-in. How cool is that? GitHub loves us - we're one of the fastest-growing libraries there, and we're used by little-known companies like Google, Waymo, and Intel. No big deal. Our founding team hails from places like Princeton, Stanford, Google, and Tesla, and we're backed by Y Combinator & other Silicon Valley heavyweights. Activeloop is hiring, and we want you! Check out our open roles on our YC page and join the fun. 10-min demo: https://activeloop.wistia.com/medias/aibvo0dst2 Whitepaper: https://www.deeplake.ai/whitepaper

Archil (fka Regatta Storage) transforms S3 buckets into a 30x faster, unlimited, local disk. Archil enables AI, analytics, and serverless applications to instantly access massive data sets without waiting for data transfer. Researchers use Archil for shareable, local storage of data set and model versions that never runs out of capacity.

Blockscope is a Web3 data platform that enables anyone interested in on-chain data to quickly spin up live or historical data pipelines (for blocks, event, transaction, wallets, token, NFTs, etc). Similar to AWS, users have access to variety of managed software services to analyze, interact with or inject captured data into their apps. Our platform is built around the core idea of off-loading the on-chain data that our customers need for their Web3 apps, and reliably storing them for quick access/use. We offer products on top of the core data we index for our customers. Some of these services are contract analytics, wallet profiling, events & transactions database as a service, push notifications, webhooks, and developer tools. Our mission is to enable everyone to easily access and consume on-chain data!

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.

Mem0 provides a memory layer for LLM applications, enabling them to remember and learn from user interactions over time.

assistant-ui helps frontend developers add AI chat to their apps. Our product is used by several dozen companies, like LangChain, Stack AI, Browser Use, and Athena Intelligence. We're building an open source Typescript/React library and a backend as a service, with first class integrations into popular agent frameworks like LangGraph and Vercel AI SDK.

Roe AI is the next-generation AI-powered data warehouse to store, process, and query unstructured data like documents, websites, images, videos, and audio. By making the impossible data possible, we help data teams become strategic enablers. Data lies at the heart of strategic decision-making, steering enterprises toward their KPIs. Roe AI accelerates these successes by providing intuitive and intelligent multi-modal data extraction, data classification & multi-modal RAG via Roe's SQL engine. Book us a call to discover how Roe AI can take your enterprise's data intelligence to the next level https://cal.com/team/roe-ai/hi-roe The future of data science is here.

Cedana (YC S23) brings hyperscaler and frontier-lab orchestration capabilities for AI workflows. Our core capability is live migration for CPUs and GPUs workloads. This increases cost savings up to 80%, accelerates time to first token 2-10x, and enables stateful reliability of training jobs even through catastrophic GPU failures. We've integrated our solution into K8s, and support Kueue and Slurm for training distributed jobs, and Kserve for serving inference.   OpenAI, Meta and Microsoft have flavors of these capabilities internally and we’re bringing them to everyone.  Our vision is to transform cloud compute into a real-time, arbitraged commodity.  https://www.cedana.ai

Bitmovin has been a first mover in almost every significant development in online video, from building and deploying the world’s first (and fastest) commercial adaptive streaming (MPEG-DASH/HLS) HTML5 Player, to being the first to achieve 100x realtime encoding speeds in the cloud. Bitmovin provides HEVC as well as VP9 live streaming with 60FPS and 4K resolution, and built the first containerized video encoding solution with Docker and Kubernetes. Bitmovin products are completely in-house developed, easy and fast to integrate and highly customizable. In combination with our great support, documentation and SLAs, this is a true enterprise offering. To find out more about Bitmovin’s video infrastructure solutions, or about any individual products, contact sales@bitmovin.com, or visit our website: bitmovin.com

We are on a mission to build the world’s most accurate search engine over complex and unstructured documents. Most AI products - whether Q&A bots or AI agents - depend on retrieval systems to provide relevant context from knowledge bases. Yet, the vast majority of these systems rely on basic semantic or hybrid search methods, which still frequently fail. These mistakes lead to inaccurate responses and hallucinations by LLMs, frustrating developers and end users alike. That is why we’re building ZeroEntropy: to add intelligence to retrieval and empower developers to create AI products that are more reliable and accurate.

QuestDB is a high-performance, open-source database for time series data. It focuses on fast ingest, blazingly fast queries and superior developer experience with SQL and open formats. QuestDB is designed from the ground up to operate as close to the hardware as possible -- eliminating ingestion bottlenecks and lowering overall hardware requirements. QuestDB powers the infrastructure for companies in Financial Services, Crypto/Web3, Space Exploration, and Energy/Renewables.

VectorShift is an AI automations platform. Teams leverage AI through VectorShift’s no-code or SDK interfaces to search through knowledge bases, generate documents, and deploy chatbots and assistants. The no-code platform gives end-users an interface to access pre-built workflows for end use cases (e.g., responding to RFPs, writing proposals, writing account plans) while letting them customize and build new ones with drag and drop components.

SST is an open-source framework that makes it easy to build modern full-stack applications on your own infrastructure.

Miru makes it easy for robotics teams to manage, version, and deploy robot configurations. Developers use our well-documented APIs and UI components to automate configuration updates for both engineers and operators. In the long term, we're building software infrastructure to accelerate the automation of the physical world.

Managing enterprise software deployments shouldn’t feel like a balancing act. At Glasskube, we make deploying, managing, and updating software in on-premises, VPC, and air-gapped environments easier than ever. Our Mission: We empower software vendors to deliver exceptional customer experiences with less complexity. Whether you’re onboarding new clients, rolling out updates, or troubleshooting issues, our Software Distribution Platform streamlines every step. The founders have been best friends for over 15 years, lived together in college, and even had a startup together. Philip was the CTO of a European Kubernetes-based company that was acquired in 2022.

The leading platform for building and managing AI voice agents that revolutionize contact center operations.

Nobody should have to employ expensive imaging scientists and engineers for their business to get the most out of their visual media. This is where imgix has your back. imgix is a platform for managing, processing, rendering, optimizing, and delivering your existing visual assets so that you can meet the demands of your business and your team can stay focused on what they do best. The business demands on web visual media can blindside even the most hyper-aware content teams and developers. Just to be competitive on the Internet, you need your images to be sized perfectly to every device, served in one of any of a half dozen web formats, optimized to the absolute best perceptual quality while maintaining pinpoint performance, and distributed to wherever your users are, anywhere in the world. And that is just the baseline. Next-level imaging concerns like copyright, licensing, accessibility, moderation, illegal or explicit content, "fake news", and more, can threaten your business at any moment if not handled properly. Get any of these things wrong and you are shedding revenue, ceding ground to competitors, and potentially exposing your business to extreme liability. By working with imgix's powerful APIs, SDKs, user interfaces, and integrations, you can rest assured that any demands on your images and videos will always be met, now and into the future, without you having to invest in building it yourself or patching together a mishmash of open source and third-party technologies. Our platform is designed to ensure that your images are always working for your business and never against it. imgix was launched through the famed Y Combinator accelerator program. Its customers include Google, Lyft, Kickstarter, AirBnb, Coursera, Bonobos, Spotify, Vimeo, Bustle, and many others.

AfterQuery is a research lab (backed by Y Combinator) investigating the boundaries and capabilities of artificial intelligence through novel datasets and experimentation. We believe that pushing the frontiers of AI requires both rigorous scientific inquiry and exceptional human data.

You add a few decorators to your agent code, and run your agent on a dataset of questions you've generated / collected in the wild. We give you back a dataset of message triplets that's ready to be uploaded to your fine-tuning provider of choice (OpenAI, OpenPipe, Anthropic, Gemini, etc) that will give you a language model ready to steer your agent in production more effectively.

ParaQuery is a fully-managed GPU-accelerated Spark solution providing double the performance at half the cost compared to solutions like Databricks and BigQuery. It is fully Spark-compatible, cloud agnostic, and has seamless integrations everywhere, all without vendor lock-in. ParaQuery makes it trivial to run large SQL and Spark workloads efficiently and serverlessly, without any data migrations.

Octopipe is ETL reimagined powered by AI that changes how companies create, maintain and scale data pipelines.

Fibers helps developers ship integrations 10x faster by syncing data from 3rd party APIs with their production databases.

At Okteto, we are working hard at creating the next-generation development experience for teams of all sizes. Have you ever seen how long it takes a developer to set up everything they need before working on a feature? We are fixing that. We want developers to focus on building value for their teams instead of wasting time setting up local environments or trying to debug "production only" issues.

Dojah is the anti-fraud infrastructure for the internet, helping businesses verify identities, detect fraud, and stay compliant with powerful APIs and no-code tools. Fintechs, banks, and digital businesses rely on Dojah’s solutions for seamless user onboarding, AI-driven fraud detection, KYC & KYB verification, and transaction monitoring. By leveraging government data lookups, biometric authentication, and real-time risk analysis, Dojah ensures businesses can mitigate fraud risks, improve user trust, and scale efficiently while staying compliant with regulatory standards.

Mux is an API-first video platform designed by experts to make world-class video streaming and analytics possible for every development team. Self-optimizing videos take the guesswork out of encodings, delivery, and renditions. Mux makes your video look beautiful on every device, every time. We offer two products, Mux Video and Mux Data. Mux Video is a simple API to advanced video streaming. Powered by data and designed by video experts, Mux Video makes beautiful video possible for every development team. Post a video or live stream with a single API call, and watch it within seconds. Mux Video handles the encoding, storage, and delivery. QoE and performance analytics provided by Mux Data is included by default for every Mux video stream.This real-time data is used to create the right renditions for every device and title, resulting in greater reliability and better viewer experiences. Mux Data is a Quality of Experience analytics platform that measures the performance metrics that actually matter to viewers: rebuffering, startup time, video quality, and errors. Our intuitive dashboard quickly gives users the insight to find issues and make improvements through features like A/B testing, industry-wide scores, alerts, and detailed information about every single video view. The product works out of the box on HTML5, iOS, Android, and major OTT platforms. All data is available via exports and APIs, making it easy to integrate Mux Data into other systems.

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.

Cyberdesk is an open source, self hostable fleet of virtual desktops. We expose a simple API to start and control hundreds of virtual desktops, perfect for building computer use agents that automate repetitive, boring desktop tasks (like data entry). Under the hood, we handle the provisioning of the desktops, a warm pool (for fast boot up times), and sending commands. Most importantly, we are fully open source and self-hostable on any Kubernetes cluster, which makes us perfect for agent developers automating in regulated industries such as healthcare, finance, insurance, and more.

Apoxy is the proxy platform for improving the security, observability, and usability of APIs and services at scale. Our programmable proxy allows developers to write code that handles the operational concerns of their applications rather than piles of configuration. We've built an ingress and egress traffic controller with a tool belt of extensions that solve common problems like monitoring, authentication, load balancing, data compatibility, and much more.

Golf lets you ship a production-ready MCP server without worrying about the plumbing. You just write the tools, prompts, and resources you want agents to call. Golf handles everything else - routing, auth, telemetry, error reporting, and deployment - all managed through our hosted platform.

We make it possible to build a serverless cloud platform that hosts custom user code with just APIs and no infrastructure team. For example, an open source framework can build a cloud offering that deploys their users code from Github or local dev environments in days instead of months.

PgDog is an application for sharding PostgreSQL. It understands SQL and can distribute queries automatically between databases. It's built for managed databases, like AWS RDS, and doesn't require any changes to application code or schema. In addition to sharding, PgDog is a load balancer and pooler, so it can act as a replacement for PgBouncer, RDS Proxy, and other Postgres scaling products. It brings the simplicity and performance of HTTP load balancing to the database. Built from the experience of sharding Postgres at Instacart during peak growth in 2020, PgDog is the answer to the old question: does Postgres scale? It does now.

Ubicloud is an open source cloud that can run anywhere. Our cloud services include elastic compute, block storage, virtual networking, managed Postgres, K8s, AI inference, and powerful IAM. Ubicloud provides these services on bare metal providers, such as Hetzner, Leaseweb, or AWS Bare Metal. You can self-host our software or use our managed service to reduce your cloud costs by 3-10x.

Terracotta AI is an AI-powered Infrastructure as Code PR reviewer for Terraform, OpenTofu, and CDK-TF code, making Infrastructure as Code changes more transparent, less risky, and faster. Terracotta stops your Infrastructure as Code deployments from becoming ticking time bombs every time you modify your environment with AI-driven, contextual feedback delivered directly to your pull requests.