Investigations at machine speed
An agent that has to wait five minutes for a query does not investigate — it guesses. CtrlB puts your complete operational history behind an interface fast enough for an agent to ask twenty questions in the time a data lake answers one.
Speed changes what an agent can do
Investigation is iterative. Every answer leads to more questions. Fast queries mean the agent can follow every lead — traditional lakes force it to guess after two or three.
Traditional tools
3 queries in 2 hours
Agent investigating checkout latency after a deploy.
Total: 2 hours 15 minutes
Investigation has barely started. Containment window is closing.
CtrlB
20 queries in 4 minutes
Same question — the agent pivots on every finding.
Total: ~4 minutes
Root cause mapped. Affected services isolated. Engineer handed the same evidence.
Why lakes stall agents
When each query takes half an hour, an agent cannot iterate, pivot, or pursue multiple hypotheses. Indexing at ingest flips that economics.
Problem · full scans
Traditional lakes are too slow for agents
- Minute-scale queries on months of telemetry
- Full table scans for a single IP or trace_id
- Per-query costs that punish iteration
- Sampled traces and short retention cliffs
- Agents throttled like dashboard users
Solution · index at ingest
CtrlB makes iteration actually possible
- Sub-second point search across petabytes
- Index at ingest — scan only matching files
- Reads are compute, not a metering event
- Complete unsampled history in your bucket
- Sustained agent volume without rate walls
The query never reads the lake
Most engines scan until they find the answer. CtrlB indexes as data lands, then opens only the cells that can match — so an agent can ask again without paying for a full scan.
waiting for the lake to be queryable…
What agents hit today
- Query latency measured in minutes, so an agent can afford two or three questions per investigation instead of twenty.
- Rate limits and per-query costs designed for humans clicking dashboards, not for a machine issuing sustained query volume.
- Sampled traces and 30-day retention, so the evidence the agent needs was discarded before the question was asked.
- Screen-scraping a dashboard UI, or a bespoke export pipeline that goes stale the moment the schema moves.
What CtrlB gives them
- Sub-second point search across petabytes, so an agent can branch, backtrack, and re-ask without a human waiting.
- Stateless query nodes that fan out on demand, so machine-scale query volume costs compute rather than a bigger cluster.
- Complete unsampled history on your own object storage, so the answer does not depend on what someone chose to keep hot.
- An MCP interface over the same data layer humans query, so the agent and the on-call engineer see identical evidence.
What you get
MCP interface
Agents query telemetry directly over Model Context Protocol — no dashboard scraping, no bespoke export job to maintain.
Full-text and SQL
The same data layer answers a free-text needle search and an aggregate SQL query, so agents do not need two backends.
Complete history
Years of retention on object storage pricing, with every retained day queryable at the same speed as today's data.
Sustained query volume
Concurrency is a compute decision, not a licensing tier. Agents are not throttled for asking too many questions.
Same evidence for humans
When an agent hands off to an engineer, the engineer opens the same query against the same data — no reconciliation step.
Your bucket, your data
Open Parquet in your own S3 or GCS. Point another engine at it whenever you want; there is nothing to export.
Same security agent. Same S3 data. Only the query layer changed.
View customer storyQuestions
How do agents actually connect?
Through CtrlB's MCP interface, which exposes search and SQL over the same data layer the UI uses. An agent authenticates, issues queries, and reads results without any screen scraping.
What happens to cost when an agent runs hundreds of queries?
Storage cost is object storage cost and does not move. Query cost is the compute the queries actually consumed, on stateless nodes that scale to zero afterwards — there is no per-seat or per-host multiplier on agent traffic.
Does the agent see sampled data?
No. Data is written complete to your bucket, so an agent investigating last quarter's incident reads the same fidelity of evidence as one investigating this morning's.
Can we keep our existing dashboards?
Yes. CtrlB can sit underneath the tools you already run — see extending your existing stack.
Give your agents the lake they can actually investigate.
Point a source at CtrlB, keep everything else where it is, and compare the same investigation side by side.