cd /news/ai-agents/keewanodb-stores-events-for-agent-an… · home topics ai-agents article
[ARTICLE · art-136037] src=blocksandfiles.com ↗ pub= topic=ai-agents verified=true sentiment=↑ positive

KeewanoDB stores events for agent analysis

Israeli startup Keewano has revealed $12 million in total seed funding for KeewanoDB, a database it says stores each entity's complete event sequence contiguously and in order so AI agents can read answers directly from raw data. Keewano says KeewanoDB can query a quarter of a billion events in half a second and scan 10 events per CPU instruction using SIMD, with access through MCP, SQL, REST, and APIs. The company, founded in Tel Aviv in 2024 by CEO Mark Kardashov, COO Dima Karger, CTO Pavel Bilbergal, and Chief Product Officer Vitaly Bukhovsky, prices the managed Keewano Cloud service by monthly active users rather than per event, with a free plan capped at 3 million events and 27,000 MAU per cycle.

by read3 min views1 publishedSep 21, 2026
KeewanoDB stores events for agent analysis
Image: Blocksandfiles (auto-discovered)

AI/ML

Israeli startup Keewano has a database storing data for machine reasoning at scale by AI agents, and has revealed $12 million in total seed funding.

Its CEO and co-founder Mark Kardashov said: “We were trying to use AI agents to answer business questions like ‘Which of our users are on the same path as the ones who churned last month?' You could see what happened on any dashboard, but why it happened lived somewhere the databases couldn’t reach. Our databases had been designed for humans querying tables, not machines that need to reason live, with complete context. To support agents properly, we realized we’d have to build a completely new storage format.”

Keewano says that, for decades, analytics infrastructure has been built for humans. Events are stored flat as rows, columns and precomputed aggregates; any question about the order things happened in has to be reassembled at query time, and the more event types a company keeps, the slower and pricier it gets. Most deployments capture only a few hundred distinct event types, because each type adds instrumentation, slows queries and costs more. Teams must decide in advance what's worth keeping, and discard the surrounding context.

Kardashov said: “In 1964, Japan didn’t try to make trains faster on the same track. They built a new line. Databases are at that fork now. AI agents need a fundamentally different data architecture, and we’ve built KeewanoDB as the new line.”

The design is similar in a way to a streaming log. The KeewanoDB keeps each entity’s complete event sequence together, contiguously, and in order, so an agent can read answers directly from the raw data rather than reconstruct them, allowing for much faster and more accurate queries. Because the sequence is stored contiguously and in order, KeewanoDB can scan multiple events in one CPU instruction, using CPU acceleration (SIMD) for processing multiple events (10) per instruction.

A semantic layer maps raw event names and fields to shared concepts and definitions. Agents, apps, and dashboards connect through MCP, SQL, REST, and APIs. With KeewanoDB, you can query a quarter of a billion events in half a second, returning context-ready results agents can reason over immediately.

Keewano says this means an AI agent can investigate questions a dashboard can't: why a customer churned, what sequence of behaviors led to an outcome, what distinguishes one group of users from another, or any new and complex behavioral questions no one thought to model in advance. An agent can see what happened before what, identify the patterns users have in common, and proactively surface emerging risks.

A fully managed Keewano Cloud service connects alongside an existing data warehouse or replaces it, using standard integrations. Organizations can plug-in their own AI agents or use Keewano’s. Pricing is not per-event, being based in the number of monthly active users (MAUs), so customers can capture more and query more.

There is a Keewano Free Plan in which you get up to 3 million events and a ceiling of 27,000 MAU per cycle. If you outgrow the events allowance or the MAU ceiling, your project goes read-only. The data stays and your queries and dashboards keep working, but new ingestion s until you upgrade.

Keewano company background

Keewano was founded in Tel Aviv in 2024 by four people:

  • CEO Mark Kardashov
  • COO Dima Karger
  • CTO Pavel Bilbergal
  • Chief Product Officer Vitaly Bukhovsky

Bukhovsky previously co-founded cloud test automation platform TestProject and Devalore, a boutique software developer, with Kardashov and so they were focussed on product and company-building. Both startups were acquired, TestProject in 2015 and Devalore in 2022. Karger previously ran a large studio at games developer Plarium looking after operations/GM-type work, while Bibergal was CTO there, working on game engines, performance, and multi-platform systems. Plurium was building and scaling games for millions of players.

The seed funding was led by Hetz Ventures with participation from a16z speedrun, Remagine Ventures, DIG Ventures and angel investors.

Get more Keewano information here.

── more in #ai-agents 4 stories · sorted by recency
── more on @keewano 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/keewanodb-stores-eve…] indexed:0 read:3min 2026-09-21 ·