Empirik spins out of Sequoia with $21M to predict infrastructure outages

By Carlos Montiel | Enterprise AI Specialist
Leer en español →
Published: 2026-09-07 | By: Carlos Montiel | Reading time: ~5 min

Empirik, an AI observability platform that predicts infrastructure outages before they happen, spun out of Sequoia Capital on September 1 with a $21 million seed round led by its own incubator, joined by Canapi Ventures and Alumni Ventures.

The problem it solves: the changes that break everything

Empirik was born inside Sequoia and targets a very specific problem for operations teams: most infrastructure outages don't start as a random failure — they start as a config change, a deployment, or an update that looks harmless until it cascades. Empirik's product monitors those infrastructure changes and infers their downstream impact before they turn into an incident alert.

How it works: an autonomous "traffic cop"

The company describes its own system as an autonomous "traffic cop" for infrastructure changes: it lets low-risk changes through without friction, throws up guardrails around medium-risk changes, and flags the most dangerous changes for human review before they're applied. It's a complementary layer to AI SRE platforms like Resolve and Sequoia-backed Traversal, rather than a direct competitor.

The founding team is led by CEO Kartik Chandrayana, former head of product at Quantum Metric, alongside Avon Puri and Sudheer Dhurjati, both former Sequoia IT leaders. Early customers already include S&P Global and Guardant Health, two organizations running high-volume, mission-critical infrastructure.

Empirik — round details Capital raised: $21M (seed) Investors: Sequoia Capital, Canapi Ventures, Alumni Ventures Independent launch date: Sep 1, 2026 Known customers: S&P Global, Guardant Health Category: predictive infrastructure observability / complementary AI SRE

What it means for companies already relying on AI agents

Empirik is another sign that "AI reliability" is becoming its own product category, separate from traditional observability. As more companies in Guatemala and Latin America hand operational tasks to autonomous agents, the relevant question stops being just "did the model answer correctly?" and becomes "what system change caused the agent to start failing?" Tools like Empirik aim to answer that second question before it turns into a customer-visible incident — something companies already running agents in production should start demanding from their infrastructure vendors.
Carlos Montiel
Enterprise AI Solutions Architect
LLMs, Agents & Orchestration Specialist
guatemalia.com/#contacto · info@guatemalia.com

Need to implement AI in your company?

Carlos Montiel is an enterprise AI solutions architect. He implements LLMs, Agents, RAG and orchestrators for companies across Guatemala and Latin America. Reach out for a consultation.

Contact Carlos Montiel

info@guatemalia.com