Starting over, in public
This site used to be about coaching. From here it is about founding a company and building a database at the same time: the engineering, the commercial decisions, and the parts I get wrong.
Founder & engineer · Basekick Labs
I build databases for telemetry nobody wants to throw away.
Twenty years in and around technical teams taught me the same lesson over and over: the data is easy to collect and expensive to keep. So I built Arc, an open, SQL-native time-series database that writes standard Parquet to storage you own. Now I'm building the company around it, and writing about both as it happens.
The flagship · Creator & maintainer
Arc Open, SQL-native time-series database for telemetry you need to keep.
Most time-series databases make you choose between keeping your history and affording it. Arc doesn't. It ingests through InfluxDB Line Protocol, writes standard Apache Parquet to storage you own, and queries recent and historical data through one SQL surface, with no retention cliffs, no proprietary formats, and no separate cold-data stack. One Go binary, DuckDB as the query engine, and the whole pipeline around it: ingestion, compaction, retention policies, continuous queries, auth, backup and restore.
- InfluxDB Line Protocol and Telegraf compatible: dual-write to validate before you cut over
- Open Parquet on local disk, S3, Azure, or MinIO. Your files, readable by anything
- Analytical SQL with window functions, CTEs, and joins across the full history
- Single binary. Runs on a laptop, an edge box, or a cluster
- Air-gap ready, FIPS builds available, AGPL-3.0
The company · Founder
Basekick Labs
The company behind Arc, and the platform around it.
Basekick Labs builds the data layer for telemetry that never stops arriving. Arc is the open core; around it sit Arc Enterprise for clustering, security, and support, a managed offering for teams who would rather not run the database, and Launchpad for the people who do. Built for aerospace and defense telemetry first, now running in manufacturing, energy, logistics, finance, and observability.
-
Arc Enterprise
clustering, RBAC, commercial licensing, and support
-
Arc Enterprise Managed
we run the database, you run the business
-
Migration paths from InfluxDB, ClickHouse, TimescaleDB, and QuestDB
-
Industry solutions
aerospace & defense, tactical edge, industrial IoT, energy, finance
The rest of the toolbox
Everything Arc needs around it
A database on its own is not a system. These are the pieces that make Arc operable: a console to run it from, a CLI, a migration path off InfluxDB, a Grafana datasource, plus a couple of things I built because the problem wouldn't leave me alone.
-
Launchpad
SvelteSelf-hosted web console for the Arc instances you run.
SQL console, schema explorer, logs, and monitoring, plus management for tokens, retention, alerts, continuous queries, and MQTT ingestion. Deploys next to Arc with Docker Compose.
-
Memtrace
GoLLM-agnostic memory layer for AI agents.
No embeddings, no vector database. Fast, structured, temporal memory that any model can read and write, because most agent memory problems are recall problems, not similarity problems.
-
Firn
GoWriter-agnostic Apache Iceberg table maintenance.
Compaction, snapshot expiry, and orphan-file cleanup for Iceberg tables, regardless of what wrote them. Open source.
-
tsm2arc
GoMigrate InfluxDB 1.x and 2.x into Arc, straight off disk.
Reads TSM and WAL files directly, with no running InfluxDB instance required. Checkpointed and resumable, so a multi-terabyte migration survives an interruption.
-
arcctl
GoThe Arc CLI, built for operators.
Query, write, import, and manage Arc databases from a terminal or a CI pipeline.
-
Grafana datasource
GoProduction Grafana datasource for Arc.
Apache Arrow transport with JSON fallback, full alerting support, and query editing that understands Arc SQL.
Writing
The founder journey, in public