data-observability
(29)
The Data Catalog That Learned to Answer Back
Born from LinkedIn’s metadata crisis, DataHub is turning the humble data catalog into a live map for humans and AI agents. The clever part is not collecting more data. It is remembering what the data means.
Bigeye Watches the Data Your AI Forgot to Question
Bigeye began by catching broken tables before they reached the boardroom. Now it wants to watch the agents reading them, too - a useful expansion for enterprises, and a lot of machinery for everyone else.
CastorDoc Turned the Data Catalog Into the Office Help Desk
The catalog once known as CastorDoc makes a persuasive case that governance works better when it feels like search, not paperwork. Its acquisition by Coalesce gives the friendly front door a much bigger engine room.
Secoda Put a Search Bar on the Data Swamp
Every company says its data is valuable. Secoda tackles the less glamorous problem: helping people find it, understand it, trust it, and stop asking the same analyst where the revenue table lives.
Wayne Jin Makes Invisible Software Matter
After Google Cloud, GitHub, AppDynamics, and Grafana, the computer-science graduate has taken on a peculiar marketing assignment: make companies care about the failures they cannot yet see.
Keith Alsheimer Has Spent a Career Making Complexity Behave
Before he sold software for unruly data systems, Keith Alsheimer trained soldiers, patented a visual method for teaching cleaners, and built a procurement company from scratch. The connecting idea is surprisingly consistent: make complicated work clear enough that people can act.
Vasu Sattenapalli Is Turning Data Trust Into a Product
After years inside enterprise data systems, the RightData co-founder arrived at a plain conclusion: a dashboard is only useful when people believe the number. Now he is building the controls that turn corporate data from a hopeful guess into something teams can use.
Maarten Masschelein Is Making Bad Data Someone’s Job
The Soda co-founder has spent his career turning silent data failures into visible, shared work. His method is less about dashboards than about focus, accountability, and giving every data user a seat at the table.
Maayan Salom Keeps Asking the Awkward Question: Can You Trust the Data?
From eighth-grade Football Manager spreadsheets to autonomous data agents, Elementary’s co-founder has built a career around the moment when a number makes you say, ‘This can’t be right.’
Elementary Gave Away the Data Alarm - Then 1,500 Teams Asked for the Fire Department
Elementary won developers with a free, dbt-native smoke detector. The harder - and more valuable - job was turning that foothold into a control room for data, analytics and AI.
Soda Wants Your Data to Rat On Itself Before It Wrecks the Dashboard
The idea behind Soda is almost rude in its simplicity: software gets tested before it ships, so why does data get to skip the exam? Two Belgians decided it shouldn't.
RightData’s $6 Million Bet: Make Bad Data Boring Before AI Makes It Expensive
Enterprise AI has made an old plumbing problem newly urgent: a pipeline can run perfectly while delivering the wrong answer. RightData sells a three-part control system designed to catch that failure before it reaches a dashboard, an audit, or an algorithm.
Precisely
The 1968 mainframe sorting company you have never heard of quietly became the plumbing behind 95 of the Fortune 100's data - and it is now betting the house on making that data safe for AI agents to act on.
Velum Labs
Velum Labs is a Y Combinator-backed San Francisco startup building an operating system for data quality. Its platform continuously watches a company's data stack, catches when metrics and definitions diverge, traces the root cause through query lineage, proposes automated fixes, and then writes enforceable data contracts so the same break does not recur. Founded in 2025 by Chilean-born childhood friends Benjamin Munoz-Cerro (CEO) and Alen Rubilar-Munoz (CTO), the company is already in production at regulated fintechs monitoring 200+ tables tied to more than $1B in assets.
Astronomer
Astronomer is a New York-based data infrastructure company and the commercial force behind Apache Airflow, the open-source standard for data orchestration. Its flagship product, Astro, is a fully managed, cloud-native platform that lets data teams build, run, and observe data pipelines at scale, and its Astro Observe layer extends that into data operations and observability. Astronomer employs a large share of Airflow's core committers, serves 700+ enterprise customers, and raised a $93M Series D in May 2025 to power orchestration for enterprise AI workloads.
DvSum
DvSum is a Sunnyvale, California data intelligence company founded in 2014 by Aashish Singhvi. Its cloud platform unifies data catalog, data quality, data governance and data lineage into a single automated system, and layers AI agents on top - including a 'Talk to Your Data' conversational agent (CADDI) that lets business users query their data in plain language. More recently DvSum has extended its agentic AI approach into telecom with an Active Network Intelligence stack (AURA) that helps operators detect, diagnose and resolve network issues before subscribers call.
Collate
Collate is the company behind OpenMetadata, the fast-growing open source project for unified data discovery, observability, and governance. Founded by Apache open source veterans Suresh Srinivas and Sriharsha Chintalapani, Collate layers AI agents, automation, and enterprise support on top of an open metadata 'context layer' so data teams - and the AI models they feed - can find, understand, trust, and govern their data. The company raised a $10M Series A led by Venrock in July 2025.
Upriver
Upriver is an AI-native data engineering platform that connects to a company's full data stack - Snowflake, Databricks, BigQuery, Airflow, dbt - then explores it, builds and validates pipelines, fixes them when they break, and encodes the tribal knowledge that usually lives in engineers' heads. Founded in 2024 by Talpiot graduates Ido Bronstein and Omri Lifshitz, the Tel Aviv company aims to make data infrastructure invisible so AI systems can run on a reliable data foundation without constant manual upkeep.
Ascend.io
Ascend.io is a Palo Alto-based data engineering platform that uses AI agents to automate the building, running, and maintenance of data pipelines across Snowflake, Databricks, BigQuery and beyond. Founded in 2015 by ex-Googler and Ooyala co-founder Sean Knapp, the company replaces hand-coded ETL plumbing with a metadata-driven automation engine and DataOps Agents that can write code, fix incidents and tune performance on their own.
Unravel Data
Unravel Data is a Mountain View, California company building an AI-native data observability and optimization platform. It plugs into modern data stacks - Databricks, Snowflake, Google BigQuery and Cloudera - and uses AI, machine learning and a context graph of workloads, infrastructure and users to automatically troubleshoot, tune and control the cost of data pipelines. Where most monitoring tools stop at telling teams what is wrong, Unravel aims to fix it: its newest engine, Arvix AI, rewrites code, reconfigures infrastructure and validates the changes before deploying them, so data engineers can build instead of firefight.
Lightup Data
Lightup Data is a Mountain View-based enterprise data observability company that lets organizations monitor data quality at scale without moving their data. Its no-code platform pushes computation down into existing warehouses and lakehouses like Snowflake and Databricks, using AI-powered anomaly detection to catch data drift, outages, and discrepancies before they reach dashboards or AI models. Founded in 2019 and backed by Andreessen Horowitz, Lightup serves enterprises including AMD, Skechers, and KFC.
Barr Moses
Barr Moses is the co-founder and CEO of Monte Carlo, the company that created the data observability category and built one of the first end-to-end platforms for catching what she named 'data downtime' - the stretches when data is wrong, missing, or simply not trustworthy. A Stanford-trained mathematician and former Israeli Air Force intelligence data unit commander, she went from Bain consulting and a VP role at Gainsight to building a data reliability company valued at $1.6 billion. She co-authored O'Reilly's 'Data Quality Fundamentals' and is one of the most cited voices on why enterprises should treat broken data like broken software.
Monte Carlo
Monte Carlo is the data and AI observability company that pioneered the 'data observability' category. Its platform helps enterprises detect, resolve, and prevent data and AI quality issues - what the company calls 'data downtime' - by monitoring data pipelines end to end, mapping lineage, surfacing anomalies, and tracing root causes. Founded in 2019 by Barr Moses and Lior Gavish, Monte Carlo serves 400+ enterprises including PepsiCo, Cisco, Nasdaq, and Comcast, and has raised $236M at a valuation of roughly $1.6B.
Mozart Data
Mozart Data is a San Francisco-based startup that gives companies an out-of-the-box modern data stack - a managed Snowflake warehouse, no-code ETL connectors, a SQL transformation layer, and built-in observability - so non-engineers can go from siloed, messy data to analysis-ready in about an hour. Founded in 2020 by Peter Fishman and Dan Silberman and backed by Y Combinator, it lets startups and SMBs skip the months of plumbing usually needed to stand up data infrastructure.
Manu Bansal
Manu Bansal is co-founder and CEO of Lightup, an AI-powered data quality and observability platform for the enterprise data stack. A Stanford EE PhD and IIT Kanpur alum, he first co-founded Uhana, a machine-learning analytics company for mobile carriers that VMware acquired in 2019. The grind of debugging bad data at Uhana became the seed for Lightup, which now runs hundreds of thousands of daily data-quality checks across petabytes of customer data and is backed by Andreessen Horowitz.
Nominal
Nominal builds the software stack that lets hardware engineering teams test, validate, and operate complex physical systems - fighter jets, nuclear reactors, satellites, rockets, robots - as fast as software teams ship code. Its platform unifies high-frequency telemetry, logs, video, and simulation data so teams can analyze and trust every test. Founded in 2022 by veterans of the U.S. Navy, Anduril, Palantir, and Lockheed Martin, Nominal reached a $1 billion valuation in 2026 and now counts four of the five largest defense contractors among its customers.
Acceldata
Acceldata is an enterprise data observability and agentic data management company that helps large organizations monitor, govern, and optimize the pipelines, warehouses, and lakes powering modern analytics and AI. Founded in 2018 by former Hortonworks engineers, it now serves customers like Oracle, Verisk, PhonePe, and Dun & Bradstreet, and recently launched an Autonomous Data & AI Platform built around its xLake Model Context Protocol for agentic workloads.
Rohit Choudhary
Rohit Choudhary is the Founder and CEO of Acceldata, the company credited with coining the term 'data observability' and building the world's first enterprise data observability platform. A career data engineer who spent years at InMobi and Hortonworks before striking out on his own in 2018, he has raised over $105 million and built a platform used by Fortune 500 companies to monitor, manage, and ensure reliability of their data pipelines in an era where bad data feeding AI models is no longer just an inconvenience — it's an enterprise liability.
Syncari
Syncari is an Agentic Master Data Management (MDM) platform built for the AI era. It unifies, governs, and synchronizes customer data in real-time across enterprise systems, producing AI-ready pipelines that fuel LLMs, copilots, and autonomous agents. Founded by ex-Marketo executives, the company is trusted by Fortune 1000 enterprises and recently crossed 2 trillion data operations.