SYSTEM Cited by 1 source
FilmLight API (FLAPI)¶
FLAPI is the API surface of FilmLight's color-science + image-processing engine — the same engine that powers FilmLight's Baselight and Daylight products used across the film-and-TV industry for color grading, dailies, and transcoding. FLAPI exposes that engine as a backend-callable API suitable for integration into automated pipelines.
Netflix's Media Production Suite (MPS) uses FLAPI as the core studio media- processing engine inside Netflix's cloud compute infrastructure (Source: sources/2026-04-24-netflix-scaling-camera-file-processing-at-netflix).
Why Netflix chose to integrate rather than build¶
Netflix's explicit framing: "building a world-class image processing engine in-house is a significant, long-term commitment: one that would require deep, continuous collaboration with camera manufacturers and the wider industry." FilmLight already has that relationship and that engine. Canonical instance of patterns/industry-api-partner-as-media-engine.
What FLAPI does in the Netflix pipeline¶
Two load-bearing roles inside MPS:
1. Camera-metadata inspection at ingest. After media is uploaded to Content Hub with ASC MHL manifests validating completeness and integrity, Footage Ingest calls FLAPI to:
- Gather camera metadata from the original camera files (OCF) across the ecosystem of supported cameras/recording formats.
- Conform the workflow-critical fields to Netflix's normalized schema (concepts/camera-metadata-normalization).
- Make the metadata searchable and reusable across the downstream pipeline (reel/timing-based matching, debugging, per-stage validation).
2. VFX-plate / deliverables generation. For visual-effects and finishing workflows:
- Debayer OCF with the correct format-specific decoding parameters.
- Crop and de-squeeze using ASC FDL (Framing Decision Lists) to preserve spatial creative decisions.
- Apply ACES Metadata Files (AMF) for repeatable color pipelines from dailies through finishing.
- Generate an array of deliverables in varied formats.
Netflix ships AMFs alongside OpenEXR deliverables so recipients know exactly which color transforms are already applied and which remain to match dailies.
Why FLAPI fits Netflix's cloud-compute model¶
The 2026-04-24 post enumerates the runtime requirements FLAPI satisfies — effectively the checklist for any tool Netflix wants to run inside Cosmos as Stratum Functions:
- Packageable as Serverless Functions in Linux Docker images — quickly invoked to process a single unit of work and shut down on completion (patterns/serverless-function-for-media-processing).
- Runs on CPU-only instances — taps Netflix's wide AWS encoding pool rather than competing for scarce GPU capacity (concepts/cpu-only-media-processing). FLAPI also supports GPU rendering; Netflix deliberately chose CPU.
- Headless invocation via Java, Python, or CLI (concepts/headless-api-invocation).
- Operates statelessly — on failure, terminate and re-launch the worker.
Netflix's implementation: bundle FLAPI in an Ubuntu-based Docker image with Java/Python glue, build Cosmos Stratum Functions that accept an input clip + output location + parameters (frame ranges, AMF, FDL), and invoke them per clip / per sub-segment.
Consistency across cloud and on-prem¶
Because FLAPI packages cleanly in a Docker image, Netflix deploys almost identical code to both:
- AWS cloud compute.
- Netflix's production compute + storage centres around the world.
This ensures consistent assessment of footage wherever it physically lives — a necessary property when productions are global and some OCF never leaves the region it was shot in.
Collaboration posture¶
FilmLight is a trusted technology partner, not a vendor Netflix consumes at arm's length. The 2026-04-24 post documents the collaboration surface:
- Roadmap alignment on new camera formats + open standards.
- Accuracy + performance validation of key operations.
- Debugging edge cases surfaced by large-scale real-world workloads.
- Evolution of the API to serve both Netflix and the wider industry.
- Joint feedback into open standards (ACES, ASC FDL).
Worked example: ACES 2 support — FilmLight provided a roadmap; Netflix engineering teams collaborated on integration; during that integration, Netflix fed integration-challenge feedback back to the ACES technical leadership.
Workflow specialists also use Baselight on their workstations to manually validate pipeline decisions before principal photography — a direct consequence of Netflix using the same engine on the backend and at the desktop.
Scope limits of this page¶
The 2026-04-24 post does not publish:
- FLAPI's wire protocol / SDK surface.
- Pricing or licensing terms.
- Performance numbers (frames/sec, latency, throughput).
- Which FLAPI versions Netflix is on.
- The full set of cameras + recording formats supported.
- Whether FLAPI is used by any other organisation at Netflix's scale, or whether Netflix is FLAPI's largest consumer.
Expand this page if FilmLight or Netflix publishes internals.
Seen in¶
- sources/2026-04-24-netflix-scaling-camera-file-processing-at-netflix — canonical source; introduces FLAPI as the core studio media-processing engine inside MPS and enumerates the two roles (inspection + deliverables) and the runtime requirements FLAPI satisfies for Netflix's cloud-compute infrastructure.
Related¶
- Parent product family: systems/filmlight-baselight
- Netflix consumers: systems/netflix-media-production-suite · systems/netflix-footage-ingest
- Netflix compute substrate: systems/netflix-cosmos
- Concepts: concepts/camera-metadata-normalization · concepts/headless-api-invocation · concepts/cpu-only-media-processing · concepts/open-media-standards
- Patterns: patterns/industry-api-partner-as-media-engine · patterns/serverless-function-for-media-processing
- Company: companies/netflix