TwelveLabs Tech Stack

Enterprise video intelligence platform powered by deep learning

Software Development San Francisco, California 11–50 employees Founded 2021 Privately Held

TwelveLabs builds a video understanding platform for enterprises, with an ML-first stack spanning PyTorch, TensorFlow, JAX, and NVIDIA inference tooling. The hiring profile is heavily engineering-skewed (49 of 83 roles), with senior and staff-level positions dominating, suggesting they're scaling production ML systems rather than selling. Active projects around multi-tenant architecture, real-time APIs, and petabyte-scale data pipelines, paired with pain points in FedRAMP compliance and low-latency distributed inference, point to a company transitioning from prototype to handling enterprise workloads.

Tech Stack 50 technologies

Core StackPython PyTorch TensorFlow FastAPI PostgreSQL Go Kubernetes Apache Spark Apache Airflow Prefect AWS BigQuery JAX Ray DeepSpeed FFmpeg Amazon Bedrock NVIDIA TensorRT ONNX Triton Inference Server OpenAPI SQLAlchemy Pydantic SQL Git CI/CD Dask GCP Azure+20 more

AdoptingKubernetes

What TwelveLabs Is Building

◆Challenges

Prototype to production transition
Scaling ml systems to petabyte scale
Security compliance
Performance and security optimization
Low latency high throughput
Performance and reliability for distributed pipelines
Meeting fedramp compliance
Cost efficiency
Scaling petabyte-scale data pipelines
Complex technical challenges

▲Active Projects

Multi-tenant architecture design and operation
Terraform automation and ci/cd pipeline development
Rodeo
Video processing backend services
Ml model inference deployment
Real-time api development
Design experience layer for developer platform and apis
Performance testing with locust
Automatic dataset generation system for vlm/llm
High-precision preprocessing engine

Hiring Activity

Accelerating85 roles · 45 in 30d

Department

Engineering

Sales

Marketing

Security

Data

Research

Design

Ops

Seniority

Senior

Staff

Mid

Lead

Director

Manager

Junior

Principal

Company intelligence

Find more companies like TwelveLabs by tech stack, pain points and active projects

Get started free

About TwelveLabs

TwelveLabs develops a video intelligence platform for enterprises, delivering machine-learning-based video analysis at scale. The product sits on a modern ML infrastructure: PyTorch and TensorFlow for training, NVIDIA TensorRT and Triton for inference serving, Ray and Spark for distributed processing, and Kubernetes for orchestration across AWS, GCP, and Azure. The company is actively scaling multi-tenant infrastructure, building real-time APIs, and automating video preprocessing pipelines to handle petabyte-scale datasets. Founded in 2021 and headquartered in San Francisco, TwelveLabs operates a 50-person team with hiring underway across the United States, United Kingdom, South Korea, and Peru.

HeadquartersSan Francisco, California

Company Size11–50 employees

Founded2021

Hiring MarketsSouth Korea, United States, United Kingdom, Peru

Frequently Asked Questions

What machine learning frameworks does TwelveLabs use?

TwelveLabs uses PyTorch, TensorFlow, and JAX for model development, with NVIDIA TensorRT, Triton Inference Server, and ONNX for production inference serving.

What cloud platforms does TwelveLabs use?

TwelveLabs deploys on AWS, GCP, and Azure, with workloads orchestrated via Kubernetes and data processing via Apache Spark and Dask.