AIOps Startups funded by Y Combinator (YC) 2026

Arga handles everything after the coding agent. Connect any platform (Slack, Grafana, Jira, etc.) and it validates AI code changes by running relevant tests. With Arga, anyone can ship, maintain, and fix production code reliably without manual testing.

AI SRE agents that automatically learn each customer’s system so they work just like an in-house engineer.

Terminal Use is an orchestration platform for background agents. Purpose-built for agents that use filesystems. Our platform is CLI-first, making it easy for your coding agents to experiment and continuously improve your deployed agents.

Sentrial is the monitoring platform for AI products in production. When your AI product/agent loops indefinitely, hallucinates, or frustrates users, Sentrial detects it in real-time, diagnoses the root cause, and recommends specific fixes. We provide the monitoring, evaluation, and debugging infrastructure that traditional tools weren't built for, so teams iterate based on production reality, not just offline evals.

Get your personal OpenClaw instance on the cloud and connected to your Slack or other messaging app in 3 minutes. Pre-installed security features and common skills.

Cumulus Labs is a serverless GPU cloud with a proprietary inference stack, purpose-built for AI teams who want faster performance, lower costs, and zero infrastructure work. If your team uses GPUs — for inference, training, fine-tuning, or any compute workload — Cumulus replaces the painful parts of your stack with a platform that just works. Most teams today are stuck choosing between bad options. Self-hosting inference means wrestling with vLLM or SGLang configurations, debugging CUDA issues, and babysitting infrastructure that breaks at scale. Managed API providers like OpenRouter or Fireworks are convenient but expensive, and you're paying their margin on top of the compute. Serverless GPU platforms give you flexibility but hit you with slow cold starts, idle billing, and no help on the inference layer — you're still on your own to make models run fast. Cumulus eliminates that tradeoff. The platform assigns containers in seconds, scales to zero when you're idle, and scales up to as many instances as you need with no waitlists or approvals. You're billed by the second for exactly the compute you use — nothing more. For inference, Cumulus ships Ion — a proprietary engine that supports all major LLMs, VLMs, and MoE architectures out of the box. Ion is optimized for latency and throughput beyond what teams typically achieve managing vLLM or SGLang themselves, with zero configuration required. Whether you're serving your own fine-tuned model or hosting an open-source model like Llama or Mixtral, Ion handles the performance layer so your team doesn't have to. Cumulus also supports checkpointing, model compilation, and LoRA serving, and the team forward-deploys custom optimizations directly for customers. For training, fine-tuning, and general container workloads, teams bring any job to Cumulus and run it on the same serverless infrastructure — no cluster management, no GPU debugging, no orchestration setup. If you're paying for inference APIs, Cumulus lets you run the same models yourself for less. If you're self-hosting, Cumulus makes your models faster without the operational burden. If you need GPUs for any workload at all, Cumulus is the simplest and most performant way to get them.

Corvera (YC W26) is the last ops hire CPG brands will ever make. We help brands increase profits by up to 40% with an AI workforce that handles their day-to-day ops. We're already saving fast-growing retail brands hundreds of hours a week, helping them serve more customers, and accelerating their growth. With Corvera’s AI agents, brands can: - Process orders end-to-end without lifting a finger - Forecast demand in real time - Automatically process purchase orders and prevent stockouts - And much more. We're founded by Chris (2x Founder; ex-CEO at Better Nature; Forbes 30U30), Dirk (ex-Google Data & AI Lead), and Matthew (ex-Head of Product at Rosemark; Princeton MEng in CS). In our first month full-time, we’ve raised $2M and secured rapid commercial traction with 15 design partners and a growing pipeline of 30+ brands. Our mission is bold: to give CPG brands hands-free supply chains that run themselves through Corvera, their go-to AI-native operating system. Visit www.corvera.ai to learn more and schedule a demo!

Agentic Fabriq is the single control plane for data permissioning throughout your entire organization across your AI and employees. Through Fabriq, any employee can tap into AI/agents with exactly the data and tools they're cleared to use by an admin. Fabriq can be used internally with employees or externally for customers while handling the auth + integrations + permissions securely for you.

Kestrel is an AI-native cloud incident response platform that brings self-healing to cloud infrastructure. It monitors cloud and kubernetes infrastructure 24/7, finds root causes, and generates exact fixes - so incidents get resolved in seconds instead of hours.

Adaptional is the first system of intelligence for insurance. Existing insurance systems organize work. Adaptional AI actually *does* the work. Our first product Underwriter AI pre processes every submission for an underwriter. AI extracts data from submission files, validates each data point using public sources, and summarizes risk using your underwriting guidelines.

Our mission is to build self-improving AI agents. We're here to close the last mile reliability gap. Through calibrated llm judges and automatic evaluations, we help companies optimize their agents 10x faster than doing manual experimentation.

Okibi is a web app that lets you create AI coworkers using natural language. It's like Lovable for creating AI agents that talk to your internal software Corporate users lose 2 hours every day doing tasks that can be automated with agents But existing agent builders have janky drag and drop UIs that don’t map to what they want exactiy. Okibi lets everyone use natural language to build the exact agent they want.

Alter is a zero-trust identity and access control platform purpose-built for AI agents. It wraps every tool call in strong authentication, fine-grained authorization, and real-time guardrails, so agents can move fast without breaking things. Each request is verified at the parameter level, authorized against granular policies, executed with least-privilege access, and fully audited in real time. Unsafe actions, whether it’s a rogue DROP TABLE or a payment above policy limits, are blocked before they touch production. Behind the scenes, Alter manages credentials, issuing ephemeral, scope-narrowed access for every interaction, then rotating or expiring it in seconds. The result: no long-lived secrets, no blind spots, and no surprises in audit. With Alter, teams can move fast on AI agent initiatives while staying fully compliant with SOC 2, HIPAA, GDPR, and internal security standards. A CISO-ready dashboard delivers real-time visibility, detailed audit logs, and compliance-ready controls, removing silos, eliminating excessive permissions, and providing complete oversight of every agent workflow.

Luminal builds an ML framework and compiler that generates GPU code. Our stack 10x's model speed while simplifying deployment and cutting idle GPU costs Github: https://github.com/luminal-ai/luminal Discord: https://discord.gg/APjuwHAbGy

Janus automates AI evaluations by using high-fidelity simulation environments, catching failures in reasoning, compliance, tool usage, and performance. The resulting datasets benchmark products and feed post-training loops to continuously improve performance over time.

We are building the data layer for AI, starting with audio. We start with data collection, curating our own proprietary set of diverse conversational data covering a wide range of languages, dialects and accents. We then leverage human expert audio annotators and our own annotation platform to process audio data for Automatic Speech Recognition. With human level transcription and diarization, our data help push the audio model frontier. Today we have over millions of hours of conversational data, and growing. If you need audio data for training or evaluating your voice models or voice agents, reach out! We offer flexible licensing deals that work for startups and enterprises, with minimal process. Audio data should besimple :)

The LLM Data Company is training models in data scarce verticals

A fully open-source, performant and actionable Weights & Biases alternative that saves you money. Check out our story here! https://www.reddit.com/r/mlopai/comments/1kkc9jp/mlopai_an_efficient_free_and_opensource/

Our first product, Stillwind Search, makes finding an electronic part as easy as writing a text message. The search engine turns natural language queries into fine-grained specifications that are matched against our proprietary database of millions of parts. This is the first step towards autonomous electrical engineering.

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.

Paragon is the AI growth engine for distributors and manufacturers, freeing up staff and growing revenue on autopilot.

As AI agents take on more consequential workflows, the hard part isn’t just whether they work—it’s whether they behave consistently with your company’s knowledge, policies, and expectations. Lucidic AI turns that institutional knowledge into consistent agent behavior by continuously testing, stress-simulating, and auto-optimizing agents against your real production scenarios. Lucidic ingests your real logs, edge cases, and operational rules, then uses controlled simulations, reinforcement learning, and Bayesian optimization to automatically discover failure modes, propose targeted fixes, and verify improvements before anything reaches production. Instead of relying on manual prompt fiddling or guesswork, your agents get a continuous improvement loop: they’re tested, corrected, and optimized based on what your business actually requires—not what a generic model assumes. The result is AI agents that reliably follow your domain logic, adapt to changes, and stay aligned across clients, configurations, and environments—without you needing to hand-engineer every prompt or behavior.

Hyrex is a task orchestration framework that uses PostgreSQL (or Hyrex Cloud) as a durable task queue. Define tasks in Python or TypeScript, send them to be processed asynchronously, and let Hyrex handle the rest. Hyrex tasks are COLD: Controllable, Observable, Large-Scale, and Durable!

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.

Sage is a HIPAA compliant, AI-native communication platform that makes home care agency client-intake handling smarter, faster, and more reliable. Teams that use Sage can save over 100 minutes of busywork per prospect, freeing up their teams to manage a larger, faster growing business.

Laminar is open-source observability for AI agents. Trace complex workflows, replay and debug agent runs, and detect anomalies across trajectories at scale.

Domu automates debt-collection calls, texts, and emails for financial services and healthcare. We work with some of the largest financial institutions in the world and win by doing what nobody else is willing to do. We are hiring :)

Elevate is bringing AI to private equity-backed roll-ups. For example, one of our customers is Guardian Restoration Partners, which is backed by Alpine Investors, the leading roll-up focused private equity firm. In three months, we integrated 10 acquisitions and enabled them to pilot AI agents built on top of our data warehouse. We know this problem personally. Before Elevate, Scott was a Vice President at a $15B private equity firm where he was leading a roll-up. Steve and Jeff are both software engineers with backgrounds across finance, big tech, and start-ups. The founding team has been friends for over a decade from their Duke undergraduate days.

Pipeshift helps engineering teams run real-time inference in production. We offer optimized runtimes to hit latency/throughput SLAs, paired with infrastructure orchestration that auto-scales and routes workloads across clusters and regions.

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.

AI Agents have arrived. They are powerful enough to read, plan, and act across your everyday life and work. The question is no longer whether they can run canned demos to check the weather, it’s whether real teams can rely on them for critical work. We optimize for adoption over spectacle: useful, accountable systems that help top professional complete real tasks. Wordware is backed by Spark Capital, Felicis, and Y Combinator, with a $30M seed round. Our beautiful, historic San Francisco office sits in the shadow of the Golden Gate Bridge, it’s full of plants and radiates innovation and calm. We genuinely love working together. At lunch we talk about alignment and AI consciousness, we have surfboards in the office ready to use, we take walking meetings on the beach and hang out after work at the bar or climbing wall. We work really hard because we care deeply about what we do. Sauna is Wordware’s AI workspace for professionals. Think ‘Cursor for Knowledge Work’ and you’re halfway there. With explicit permission, it connects to tools like Gmail, Calendar, Slack, and thousands more, to gather the right context to move your work forward. It learns who you are, drafts and refines your documents, presentations, and datasets, researches across the web and your org’s internal knowledge, summarizes long threads to cut through the noise, proposes next steps, even acts on your behalf when enabled. You can work with Sauna live, side-by-side, or schedule jobs to run automatically in the background, with clear permissions and easy off-switches. Sauna brings the power of AI Agents to all professionals.

Respan (formerly Keywords AI) gives teams a unified control plane to trace and evaluate agent behavior without guesswork, automatically surface issues, and fix what breaks faster. Respan connects production observability, automated and human-in-the-loop evaluations, and an adaptive AI gateway to close the loop between detection, decision, and action - so agents improve continuously in production. Respan is trusted by 100+ AI startups and enterprise teams. Today, the platform process 1B+ logs and 2T+ tokens every month, supporting more than 6.5M end users.

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.

Find and fix your agent’s most critical failures in hours, not days. Atla helps developers cut time spent on manually reviewing traces. Atla’s LLM judge evaluates your agent step-by-step, uncovers error patterns across runs, and suggests specific fixes—so you know exactly what to fix and why. Atla supports the most popular agent frameworks teams build with, including LangChain, CrewAI, and OpenAI Agents. With real-time monitoring, automated error detection, and prompt experimentation, Atla gives teams the visibility and control needed to confidently ship agentic systems that work. We’re a team of researchers, engineers, entrepreneurs and operational leaders. Our expertise in evals was honed through training our own purpose-built LLM Judges, Selene and Selene Mini, which are available open-source and have been downloaded 60,000+ times.

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

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.

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.

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.

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.

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.

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

🚀 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!

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.

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).

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.

Openlayer is the governance platform for AI applications.