Tecton, which helps companies deploy real-time machine learning models faster with less data engineering effort, raised a $100M Series C led by Kleiner Perkins
Context & Ripple Effects
Tecton's $35M Series B in December 2020 launched its ML feature store just a week after AWS debuted SageMaker Feature Store — an early signal that real-time ML infrastructure would be contested by both startups and hyperscalers. The $100M Series C led by Kleiner Perkins, coming off a wave of MLOps fundraises like OctoML's $85M Series C, put Tecton on a path to a $900M valuation in 2022.
The arc since then validates the infrastructure thesis but not the independence one: Databricks now plans to acquire Tecton outright (the 2025 deal, undisclosed sum), folding the feature-store leader into a data platform.
First-order effects
- Kleiner Perkins' $100M check gives Tecton the capital to scale real-time model deployment against AWS's SageMaker Feature Store, which had already commoditized part of the feature-store layer within a week of Tecton's launch.
- Kleiner Perkins is deploying from freshly raised funds — including a $2.5B growth vehicle aimed explicitly at AI startups — making Tecton one of its marquee AI-infrastructure positions.
Second-order effects
- Hyperscalers' entry into feature stores forced Tecton to differentiate on real-time performance and reduced data-engineering effort rather than on the category itself, while adjacent tooling rounds (OctoML for model optimization, Tonic.ai for synthetic data) crowded the same enterprise ML stack.
- Rival data platforms watching Tecton's traction faced a build-or-buy decision on feature stores — a decision Databricks ultimately answered with an acquisition.
Third-order effects
- The pattern across this cohort — large 2021-22 fundraises followed by absorption into platforms — points to MLOps consolidating as a feature of data platforms rather than surviving as a standalone vendor category.
- For enterprises, buying ML infrastructure increasingly means committing to a platform suite, with independent point-solution vendors either exiting or being priced out of the stack.
The trend: ML infrastructure startups that raised aggressively during the 2021-22 funding cycle are being consolidated into data and AI platforms, with feature stores absorbed rather than left independent.