Staff Data Platform Engineer

Location: NYC HQ or Remote (U.S.)| Full-time | Engineering

Who we are.

Reality Defender is an award-winning cybersecurity company helping enterprises and governments detect deepfakes and AI-generated media. Utilizing a patented multi-model approach, Reality Defender is robust against the bleeding edge of generative platforms producing video, audio, imagery, and text media. Reality Defender's API-first deepfake detection platform empowers teams and developers alike to identify fraud, disinformation campaigns, and harmful deepfakes in real time.

Backed by world class investors including DCVC, Illuminate Financial, Y Combinator, Booz Allen Hamilton, IBM, Accenture, Rackhouse, and Argon VC, Reality Defender works with leading enterprise clients, financial institutions, and governments in order to ensure AI-generated media is not used for malicious purposes.

Youtube: Reality Defender Wins RSA Most Innovative Startup

The Staff Data Platform Engineer Role.

Reality Defender's detection models are only as strong as the data behind them, and every new generative tool raises the bar on the volume and variety of media we need to train and evaluate against. The Staff Data Platform Engineer will own the shared AI data platform behind our vision, audio, and multimodal research, turning acquired and internally generated image, video, audio, and text data into reliable, reproducible datasets, from continuous ingestion through processing, governance, and export. This is a senior, hands-on role where you'll complete our Databricks and distributed processing platform and build the reusable infrastructure behind our in-house synthetic data generation, including producing the very deepfakes our models learn to catch. This is a full-time role, reporting to our Director of Data. Responsibilities include:

• Own the end-to-end data platform for AI research: continuous ingestion, preprocessing, governance, and export of image, video, audio, and text data at high volume.

• Complete and scale our Databricks lakehouse and distributed processing stack (Spark, Ray, or similar), making architecture and build-vs-buy decisions that hold up as data volume grows.

• Build reusable infrastructure for in-house synthetic data generation, so researchers can produce new deepfake variants across modalities on demand instead of one-off.

• Make every dataset reproducible and traceable through versioning, lineage, metadata, quality checks, and stable train/eval splits, so any model result can be tied back to the exact data behind it.