marcmacmac

Obsessed with Performance at | @ycombinator F26

[Marc Höffl @marchoeffl]

Today we are launching @ObsessionDB, the realtime data warehouse for agents.

Your agents are hungry for data. Feeding them is getting expensive.

Your data warehouse wasn't built (or priced) for that.

We learned this the hard way. We scaled our last company to petabytes on ClickHouse and watched the infra bill go vertical.

So we built @ObsessionDB . The ClickHouse engine, rebuilt with compute and storage separated, so your infra scales with your agents instead of your bill.

100x faster and 20x cheaper than Snowflake and Databricks. Already live, processing trillions of rows a day.

Agents are just getting started...

[marcmacmac @marcmacmac]

rewrite go -> rust
adapt the changes and concepts -> go
loop ... enjoy your benchmarks

[marcmacmac @marcmacmac]

@marchoeffl ... don't make any mistakes

[marcmacmac @marcmacmac]

@samlambert @saisrirampur 🍿

[marcmacmac @marcmacmac]

what i learned so far in SF: you need to become a first name

[marcmacmac @marcmacmac]

Love your data.

[marcmacmac @marcmacmac]

agent2agent plain simple for multiple co-workers in several repos:
- mkdir agent-coms/
- ultracode + /goal
- tell claude/codex to discuss with and update the team
watch the magic happen 🍿

[marcmacmac @marcmacmac]

@XyraSinclair @andy_pavlo let's talk. we can handle the load and data growth easily, so you could focus on growth. dm me if you're interested

[marcmacmac @marcmacmac]

@shamsharoonn works... until you add an event table and a count takes longer than the coffee machine.

[marcmacmac @marcmacmac]

@bugrza The vibecoder database lifecycle:
1. Postgres for everything
2. add an "events" table
3. COUNT(*) takes 40s
4. discover columnar databases exist

We build ObsessionDB for step 4 ;)

[marcmacmac @marcmacmac]

All the countless data-gold rotting around on broken table schemas, that noone dares to migrate or make any sense out of it. With Jev you can classify it on read time - a new generation of apps/features that harness it is incoming

[Rafa Aviles @avilesrafa]

We just added Jev support to ObsessionDB/ClickHouse.

The interesting part isn’t just “LLM in SQL”, it’s that Jev is fast enough to make semantic queries feel like part of the database workflow.

You can analyze AI agent sessions on the go!

aiFilter(session, 'was the user frustrated, map('model','jev-1.13.0'))

or score frustration, success, intent, exposed credentials, etc.

At write time: turn meaning into columns.

At read time: ask new questions you never instrumented.

~50 rows/request, ~300ms in our current setup.

Feels like a new primitive for AI product telemetry.

[marcmacmac @marcmacmac]

@XyraSinclair @andy_pavlo what's you plan once your machine runs out of NVMe?

[marcmacmac @marcmacmac]

Large scan queries in a compute-storage separated database are often limited by inter-node bandwidth, not cpu or memory. ObsessionDB's locality aware distributed cache reduces this traffic to less than 10%. Less waiting, more working -> faster queries, more performance per dollar

[marcmacmac @marcmacmac]

Just waiting for the PR getting merged to finally see ObsessionDB on https://t.co/vaaDROVbc6. Spoiler: It's fast