Limited Founding offer: 20% off, locked in forever Claim
Datahyena
Enterprise Bulk delivery to your storage

The feed lands in your bucket.

Funding rounds, acquisitions and executive moves, written to your S3 bucket or SFTP host on your schedule. Date-partitioned, filtered to your scope, and incremental by default, so loading the feed is a path glob rather than an integration project.

Talk to us about delivery Read the docs

S3 · SFTP · CSV, JSON, JSONL · gzipped

your bucket 
s3://your-bucket/datahyena/
├── 2026-08-13/
│   ├── funding_events.jsonl.gz
│   ├── acquisitions.jsonl.gz
│   └── exec_moves.jsonl.gz
├── 2026-08-14/
│   ├── funding_events.jsonl.gz
│   ├── acquisitions.jsonl.gz
│   └── exec_moves.jsonl.gz
└── 2026-08-15/
    ├── funding_events.jsonl.gz
    ├── acquisitions.jsonl.gz
    └── exec_moves.jsonl.gz

One folder per run. Each file holds only what changed since that dataset last landed cleanly.

Why files

Some teams do not want an API.

If the data is going into a warehouse, an API is a middleman you have to build and then maintain. File delivery skips it. The records arrive where your pipeline already looks, in a shape your loader already understands.

Your infrastructure

Lands in your bucket

Point us at an S3 bucket or an SFTP host and the feed arrives there on schedule. Nothing to poll, nothing to host, no integration to write on your side. Credentials are encrypted at rest and never returned once stored.

S3 · SFTP · encrypted at rest
Predictable layout

One folder per run

Every delivery writes a date-partitioned folder with one file per dataset. Loading yesterday means reading yesterday, so an incremental warehouse job is a path glob rather than a diff.

date-partitioned folders
Scoped to you

Filtered before it is written

Each dataset carries its own filters, from round stage and deal size to country, industry and seniority. You receive the slice you actually bought, not the whole corpus with a note to filter it yourself.

filtered before write

Three formats, all compressed.

Pick the one your loader wants. Every file is gzipped by default, and the records carry the same fields you would get from the API, including the confidence score and the source references behind each event.

JSONL

One JSON object per line. Nested objects preserved. The default, and the easiest to stream into a warehouse.

JSON

A single array per file. Convenient when a downstream tool wants one document rather than a stream.

CSV

Flat columns with nested fields flattened to dotted headers. For teams loading straight into a spreadsheet or a legacy pipeline.

Same records, same shape

Resolved before it reaches you.

Every record arrives deduplicated and resolved to a stable company identity, with firmographics attached and a confidence score on the event itself. Anything below our serving threshold is never written at all, and every event carries the sources it was drawn from, so any row is auditable.

funding_events.jsonl 
{
  "id": "019e6a0a-4f8b-7c10-9a1d-2c1f5e77b3a2",
  "announcedAt": "2026-08-15",
  "company": {
    "name": "Galaxea AI",
    "domain": "galaxea.ai",
    "country": "JP",
    "industryGroup": "Technology, Information and Internet",
    "employeeCountBucket": "51-200"
  },
  "round": "series_b",
  "amountUsd": 145000000,
  "investors": [
    { "name": "Global Brain" },
    { "name": "JAFCO Group" }
  ],
  "signalCount": 4
}
How it is sold

Set up for you, not self-served.

File delivery is part of an enterprise agreement rather than a plan you can pick at checkout, because the destination, the filters, the cadence and the licensed scope are all configured for your use case. Pricing is an annual licence on the feed, set by which signals and geographies you need, how often it lands, and whether the rights are internal or extend to your own product.

Historical backfill at signing is included. The API, webhooks, watchlists and MCP remain available on the standard plans and work alongside file delivery.

Book a call Send us the requirement

Delivery, answered.

What teams ask before wiring it into a warehouse.

Do I get the whole database every time?
No. The first delivery can include a full historical backfill if you want one, and every delivery after that contains only what changed since that dataset last landed successfully. If one dataset fails while others succeed, only the failed one re-covers its own gap on the next run, so a partial failure never silently skips records.
Which destinations are supported?
Amazon S3 and any S3-compatible object storage, and SFTP with either password or private-key authentication. If you need a destination we do not support yet, ask, because adding one is usually straightforward.
Does this work with Snowflake, BigQuery or Databricks?
Yes. The files land in a layout warehouses ingest natively, so in Snowflake you point a Snowpipe at your bucket and every delivery loads itself as it arrives, with no schedule to maintain on your side. We provide the setup SQL. Databricks Auto Loader and BigQuery external tables follow the same pattern. If you would rather we wrote into your Snowflake tables directly, that can be part of the agreement.
How often does it run?
Daily is the common default. Weekly suits research and analysis workloads and costs less. Hourly is available where the timing genuinely matters. You choose the hours in UTC, so the feed lands before your own jobs start.
Is everything compressed?
Yes. Every format is gzipped by default, which is usually an order-of-magnitude reduction on this kind of record. You can turn it off if a downstream tool needs plain files.
How do I know a delivery actually happened?
Every run is recorded with its status, row count, byte count and duration, per dataset, and a failed run carries the reason. Nothing is written silently and nothing fails silently.
Can I still use the API and webhooks?
Yes, and most customers do both. File delivery suits bulk loading and analysis, while the API and webhooks suit real-time reaction. They read the same records, so the two stay consistent.
Is file delivery available on self-serve plans?
No. Scheduled file delivery is configured per customer as part of an enterprise agreement, because the destination, filters, cadence and scope are all set up for you rather than self-served. The API, webhooks and MCP are available on the standard plans.

Start pulling signals in minutes.

Create a key, claim your 50 free credits, and make your first request today. No sales call, no credit card.

Get your API key Read the docs

50 free credits · no credit card