Open SourceSolutionsCustomersBlog
We open-sourced FilamentExplore the docs

Building open source data infrastructure for the context era.For data engineers, by data engineers.

Backed by
Open source

Our data movement engine

Any source01

Databases, APIs, event streams, and files in object storage, all moving through one pipeline.

Sources
AirtableSource

Bases, tables, records and views from your workspace

AttioSource

CRM records, lists, notes and workspace members

BigQuerySource

Tables and query results from BigQuery datasets

GitHubSource

Repos, issues, pull requests, commits and reviews

PostgresSource

Full loads and logical replication from Postgres

MySQLSource

Tables and binlog change streams from MySQL

MongoDBSource

Collections and change streams from MongoDB

SalesforceSource

Accounts, contacts, opportunities and custom objects

HubSpotSource

Contacts, companies, deals and engagements

StripeSource

Charges, customers, subscriptions and invoices

NotionSource

Pages, databases and blocks from your workspace

SlackSource

Channels, messages, threads and users

LinearSource

Issues, projects, cycles and comments

JiraSource

Issues, sprints, boards and worklogs

Amazon S3Source

Parquet, CSV and JSON files in object storage

PostHogSource

Events, persons and feature flags

Google Analytics 4Source

Events, sessions and conversion reports

NetSuiteSource

Transactions, customers and saved searches

filament — zsh

Any sink02

Snowflake, BigQuery, ClickHouse, Iceberg, wherever your data needs to land.

Comparing Filament against the field

2.6 M row/s

Measurably faster

Rows per second calculated, outperforming Airbyte 91x on the 298 million row NYC Taxi load.

</>

Pluggable by design

State, events, and secrets run on backends you choose, swap any one without touching a pipeline.

CRC32-C

Verified integrity

Every record is hashed on read and write. Batches fail unless counts, bytes, and checksums match.

Benchmarked on identical hardware and data, tuned for peak efficiency.

View the benchmark

Move data with

We can help
Impact

Successful AI deployments run on context that's governed, access controlled, and centrally managed. We help ambitious teams build them.

We are an applied data engineering lab helping organizations own the context layer, because institutional knowledge is your most valuable asset.

We build for scale

We know volume, constraints, and the failure modes that follow

We deploy on your stack

We build alongside your team, in your repos, in your systems

We are a partner, not a vendor

No black box, no vendor lock-in, no penalty for growing

Ingestion
attionotionslackpostgresclickhousemysqlgithubstripe
granolalinearopenairamps3airtablebigqueryposthogga4
hubspotjiraclaudemongomotherduckredshiftsalesforcesnowflake
icebergsquarechargebeenetsuitegeminisapresend

Reliable data integration

Records in databases, process in tools, knowledge in docs. We pull all three into one foundation you own and keep it current with every insert, update, and delete.

Introspection
works atlives inlocated inattendedhosted by

Unified semantic modeling

We define your entities, attributes, and relationships once in a shared ontology, governed centrally, so every team and every system means the same thing.

Intelligence

Governed access and action

The same model powers your dashboards, your product, and your agentic workflows. Every action that writes back is checked against your rules before it runs.

Looking for something custom?

We partner with select companies on tailored solutions.

Let's work together
FAQ

QuestionsAnswered

What does Galaxy do?

Galaxy is an applied data engineering lab. We build open source data infrastructure, and we join your team to put it to work. Our engine, Filament, moves data from your databases, SaaS tools, and file storage into your warehouse at up to 2.6 million rows per second. Our engineers build the layer around it. One place where all your data lands, one definition of every business term, and one set of rules for who can see what. Your dashboards, your product, and your AI agents all read from the same model.

What is Filament?

Filament is Galaxy's open source data movement engine. It copies data from your databases, apps, tools, and file storage into your warehouse or lake, as a full load, just the changes, or a live stream. Every batch is verified on both ends, and a failed run picks up where it left off. It runs on your own infrastructure with nothing metered and no paid tier. In our published benchmarks it finished first against every tool we tested, using less memory and less CPU. Managed ELT charges you the most right when you succeed. Filament costs the compute you already run.

What is enterprise context management?

Enterprise context management is how a company turns everything it knows into something its people and its AI agents can trust. Your data lives in databases, your process lives in apps, and your knowledge lives in docs and in people's heads, and they rarely agree on what a customer actually is. Context management pulls all of it into one living model of the business that you control. Every term is defined once. Every record is matched across systems. Every fact carries its history. Every person and every agent sees only what they should, and any action that writes back is checked against your rules first. Get this layer right and your AI is right. Skip it and your AI is guessing.

What does working with Galaxy look like?

Galaxy is an embedded applied data engineering team. Senior engineers join your team and help you build the data stack that sets you up for what comes next. It starts with a real problem. A data bill growing faster than the business. A warehouse that needs standing up. An AI agent that answers confidently and wrong. We get the data in, give it one model the whole company agrees on, put rules around it, and ship it as pull requests your team reviews into your own cloud. Working software lands in weeks, not quarters, and your team can run it without us because they built it with us. Email hello@getgalaxy.io to talk to an engineer.

How do you handle security and compliance?

We build on your stack, so we play by your rules. Everything runs in your cloud, under your access controls, your network, and your compliance program. Your data stays in your accounts, and nothing ships without your team's review. Your security posture stays exactly what it already is. What we build is secure by default, with row-level access and audit trails from the start, so every agent stays inside the permissions you set and leaves a record you can check. You own the keys, the code, and the paper trail.

Why does Galaxy build in the open?

Because your company's knowledge is your most valuable asset, and the infrastructure that holds it should not be rented, billed by the row, or locked in a format that makes leaving expensive. That kind of pricing punishes growth. Good infrastructure should make you want to move more data, not less. So Filament is Apache 2.0, one repository, nothing gated, and every benchmark ships with the code so you can rerun it. No black box, no vendor lock-in, no penalty for growing. Galaxy was built on open source, and we return the favor by releasing each piece of our context management system as it matures. Filament is first.

Own your knowledge stack

Let's work together