Startup profile for Runta: latest funding, latest known valuation, total disclosed funding, investors, status, sources, and why the company is interesting. Part of the Venture Capital Tracker startup directory.
Venture Capital Tracker logo
Venture Capital Tracker

Startup profile · funding coverage

Runta funding, valuation and investors

Execution layer and sandboxed runtime infrastructure for production AI agents.

Keep track of Runta

Save this profile to your VCT watchlist for a quick return.

View watchlist

Funding, valuation & investors

Answer-first snapshot

Latest funding

Seed

$20M · July 2026

Latest known valuation

>$100M post-money

Seed · July 2026

Total disclosed equity funding

$20M

Excludes debt, grants, acquisitions, secondaries, and IPO proceeds.

Current status

private

seed · San Francisco, California

Investors in latest funding

Lead: Andreessen Horowitz

Sources: latest funding Last verified: 2026-07-25

Overview

Runta builds an execution layer that wraps AI agents in isolated, stateful runtimes with resource caps, network egress limits, secret management, and audit logs. Founded by Guanlan Dai (formerly Cloudflare and Kong), the San Francisco company raised a $20M seed in July 2026 led by Andreessen Horowitz's Martin Casado at a post-money valuation exceeding $100M. Runta targets teams deploying coding agents and autonomous workflows who need guardrails against file deletion, credential leakage, and runaway spend without rewriting existing agent frameworks.

Why Runta is interesting

Agent deployments need hard boundaries on OS, network, and spend — Runta containerizes agent runs with policy controls, targeting the gap between demo agents and governed production workloads.

Product & use cases

Runta provides containerized agent runtimes with CLI, Python SDK, and dashboard — enforcing CPU/memory limits, network policies, secret injection, and immutable action logs for Codex, Claude Code, and custom agents.

  • Sandbox coding agents before they touch production repos
  • Cap agent network egress and spending in enterprise environments
  • Audit every autonomous action for compliance and debugging
  • Run stateful agent sessions with managed credentials

Key facts

  • Seed (Jul 2026): $20M led by Martin Casado at a16z; post-money >$100M — Runta blog
  • Founder Guanlan Dai previously built edge infrastructure at Cloudflare and Kong
  • Angel participants include Jeff Dean, Fei-Fei Li, Ali Ghodsi, Thomas Wolf — not VCT directory funds
  • Product: CLI, Python/TypeScript SDKs, dashboard for agent runtime management

Funding history (newest first)

Seed

2026-07 $20M Valuation: >$100M post-money

Source: https://runta.com/blog/runta-the-execution-layer-for-agents/

Investors in our directory

Funds linked from Runta's profile — open a fund page for stage focus and related deal articles.

Andreessen Horowitz

Competitive landscape

Edge: Purpose-built for agent execution rather than generic VMs — policy controls and secret handling are native to the runtime, not bolted on.

Agent infrastructure is crowded with sandboxes and serverless platforms, but few products center on governed production deployment. Runta bets enterprises will pay for execution-layer controls as agents gain file-system and network access. a16z's Casado leading the round signals infra conviction.

  • E2B direct

    Cloud sandboxes for AI code execution; Runta emphasizes agent governance and production policy layers.

  • Modal adjacent

    Serverless compute for ML workloads; less agent-specific isolation and audit focus.

  • Docker / Kubernetes incumbent

    General container orchestration; requires custom wiring for agent-specific guardrails.

  • Cloudflare Workers adjacent

    Edge compute with isolation; founder overlap but different product surface.

Notable stories

  • Dai founded Runta after realizing production agents need an execution boundary — the company blog frames it as 'a stateful computer for every AI agent' with isolation built in.
  • The July 2026 seed valued Runta above $100M before broad commercial launch, reflecting investor appetite for agent infrastructure.

Industries

Infrastructure & Cloud AI & Machine Learning Cybersecurity

FAQs about Runta

Practical answers founders, operators, and investors typically search for.

Runta provides an execution layer that runs AI agents in isolated sandboxes with resource limits, network controls, secret management, and audit logs — so teams can deploy agents safely in production.
Andreessen Horowitz (/fund/andreessen-horowitz) led Runta's $20M seed in July 2026. Individual angels including Jeff Dean and Fei-Fei Li participated but are not in the VCT fund directory.
Runta announced a $20M seed round in July 2026, led by a16z's Martin Casado.
Public reporting on the July 2026 seed cited a post-money valuation exceeding $100M. Later marks not verified here.
Guanlan Dai is founder and CEO. He previously worked on edge infrastructure at Cloudflare and Kong.
Both offer sandboxed compute for AI agents. Runta emphasizes production governance — spend caps, egress limits, and audit trails — while E2B focuses on fast cloud sandboxes for code execution.
Runta is an early-stage private company; profitability has not been publicly disclosed.
Runta supports Codex CLI, Claude Code, OpenClaw, and custom agents via its CLI and SDK, per runta.com/docs.
San Francisco, California.
We track a16z-backed companies relevant to founders and investors following AI infrastructure and enterprise software.
Website All startups Funding news

By Venture Capital Tracker

Last updated:

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.