---
title: "Empirik’s $21M Seed: Sequoia Spins Out Change-Risk AI for Infrastructure Outages"
description: "Sequoia-incubated Empirik launches with $21M from Sequoia, Canapi, and Alumni Ventures — predicting outage ripple effects from infrastructure changes before alerts fire, with named customers including S&P Global."
date: 2026-09-01T00:00:00.000Z
tags: ["2026-vc-news", "startup-funding", "venture-capital", "enterprise-saas", "ai-ml", "devops"]
source: https://venturecapitaltracker.com/2026-empirik-21m-seed-sequoia
---

# Empirik’s $21M Seed: Sequoia Spins Out Change-Risk AI for Infrastructure Outages

> Sequoia-incubated Empirik launches with $21M from Sequoia, Canapi, and Alumni Ventures — predicting outage ripple effects from infrastructure changes before alerts fire, with named customers including S&P Global.

**[Sequoia](/fund/sequoia)** incubated **Empirik** inside its own IT environment from **2023**, then spun it out on **September 1, 2026** with a **$21 million** seed from Sequoia, Canapi, and Alumni Ventures (TechCrunch).

**Spine:** coding agents sped up software change volume; Empirik wants the matching **verification layer for infrastructure changes** — predict the outage before the Slack page.

## Key facts

| Field | Detail |
| --- | --- |
| Company | Empirik |
| Origin | Built by Sequoia IT leaders Avon Puri & Sudheer Dhurjati; CEO Kartik Chandrayana |
| Round | **$21M** seed |
| Lead / core | [Sequoia](/fund/sequoia) (+ Canapi, Alumni Ventures) |
| Named customers | S&P Global; Guardant Health; major CPG (unnamed) |
| Positioning | Complementary to AI SRE tools (Resolve; Sequoia-backed Traversal) per Balkansky |

## Who uses the product — and for what job

**Users:** DevOps and SRE teams drowning in change velocity from human and AI-authored deployments.

**Job:** model dependency ripple effects of proposed or recent changes; auto-allow low risk, constrain medium risk, escalate high risk — shift left from reactive monitoring.

## Why now

- LLM coding tools increase PR and deploy volume; reactive APM alerts arrive too late.
- Sequoia saw the problem in its own production environment — unusual incubator proof.
- Same-day Sequoia also led [AIR](/2026-air-security-50m-seed-sequoia-greenoaks)’s first seed: agents and infra changes both need guardrails.

## Why Sequoia — portfolio fit

Partner Bogomil Balkansky argues most observability fails to understand complex dependencies; Empirik is a dedicated **change-tracking** layer. Incubator-to-portfolio continuity is the fit story — and the conflict-of-interest diligence question buyers will ask.

| Dimension | Fit |
| --- | --- |
| Stage | Seed after multi-year incubation |
| Thesis | Verification economy around AI-sped change |
| Risk | Feature absorption by Datadog/cloud ops; false-positive fatigue |

*Canapi and Alumni Ventures are covered without inventing `/fund/` pages.*

## What remains undisclosed

- Valuation
- ARR / contract sizes
- Measured incident reduction vs control
- Exact HQ

## Competitive map

Traditional observability (Datadog, etc.) and AI SRE platforms optimize detection and remediation. Empirik claims the **pre-deploy change-risk** seat — complementary until someone bundles it.

## Takeaway

The anomaly is **incubation inside the investor**. That accelerates product-market learning and raises the bar: with Sequoia’s brand and $21M, Empirik must show outage prevention that generic AI SRE chat cannot copy.

### Sources

- [TechCrunch](https://techcrunch.com/2026/09/01/sequoia-incubated-empirik-launches-with-21m-to-predict-outages-before-they-happen/)

**By:** [Venture Capital Tracker](https://venturecapitaltracker.com/editorial-policy)

**Editorial note:** AI tools assisted with research, structure, or drafting. Venture Capital Tracker retains human editorial responsibility for factual accuracy, relevance, and source quality before publication.