Echo raises $35M to secure the enterprise cloud’s base layer with autonomous AI agents

Echo announced a $35 million Series A on December 16, 2025, led by N47, to commercialize a secure-by-design alternative to conventional container scanning. The company says it rebuilds container images, removes unnecessary components, signs the resulting artifacts, and continuously maintains them with AI agents. The round followed a $15 million seed announced in July 2025, bringing Echo’s announced funding to $50 million.

The proposition is straightforward: instead of discovering inherited vulnerabilities after an application is built, replace the vulnerable foundation with a maintained image designed to contain fewer known flaws. That could reduce security workload for large and regulated organizations, but “CVE-free” remains a scoped, time-dependent claim—not proof that an application or cloud environment is secure.

#	Preview	Product	Price
1		Nimo AI NAS, Agentic Computer Mini PC and AI Server, AMD Ryzen 7 PRO 8845HS(up to 5.1 GHZ, beat...	$1,999.99	Buy on Amazon

What Echo announced

Echo’s Series A was announced on December 16, 2025. N47 led the round, with participation from Notable Capital, Hyperwise Ventures, and SentinelOne’s S Ventures. Combined with the company’s $15 million seed announced on July 31, 2025, Echo says it has raised $50 million in roughly 10 months.

Echo’s funding announcement identifies Eilon Elhadad as CEO and Eylam Milner as CTO. The founders previously built Argon, which the company says was acquired by Aqua Security for $100 million. Echo says the new capital will support its secure software-infrastructure platform, image catalog, engineering and enterprise go-to-market; the announcement does not provide a detailed spending allocation.

#1 Best Overall

Nimo AI NAS, Agentic Computer Mini PC and AI Server, AMD Ryzen 7 PRO 8845HS(up to 5.1 GHZ, beat i5-1235u) up to 132TB ZFS Hybrid Storage, Dual 10GbE for 24hr AI Agent

• [Local AI Inference & 70B Model Ready] Equipped with the AMD Ryzen 7 PRO 8845HS processor, NEXUS is engineered for heavy local AI workloads. With a full-size GPU bay, it runs 70B LLMs natively without an internet connection. Ideal for AI developers and tech enthusiasts who need private environment for coding and model testing.
• [132TB Mass Storage with ZFS Integrity] Features a hybrid storage architecture (3×NVMe + 4×3.5" HDD) supporting up to 132TB. Utilizing the enterprise-grade ZFS file system and ECC memory, it prevents data corruption and bit rot—a must-have for professional photographers and video editors safeguarding 4K/8K RAW footage.
• [OpenClaw-Driven Automation Workflow] The built-in OpenClaw execution layer allows complex automated tasks to be processed locally. Even when offline, your backup schedules and AI file organization continue seamlessly. Say goodbye to monthly cloud subscriptions and high latency.
• [Dual 10GbE & USB4 Ultra-Connectivity] Experience server-class speeds with dual 10GbE ports and a 40Gbps USB4 interface. It enables multi-user real-time collaboration on large project files directly from the NAS, ensuring zero-lag editing for creative studios and production teams.
• [Open-Source ZimaOS for Total Privacy] Running on the fully open-source ZimaOS, NEXUS ensures your data stays physically on-premise with no backdoors. It acts as a "Digital Fortress" for privacy-conscious families and small businesses who demand absolute data sovereignty.

$1,999.99

Buy on Amazon

The same announcement names Varonis, EDB and UiPath as production customers or references. Those are company-reported customer claims, not independent performance validation.

Funding timeline

Why the container base layer matters

A container image is more than application code. It commonly bundles an operating-system userland, a language runtime such as Python or Node.js, system libraries, package dependencies, utilities and configuration inherited from upstream layers. If that foundation contains a vulnerable package, many downstream application images can inherit the same finding.

This creates a recurring shift-left problem: security teams may discover hundreds of findings only after developers have built and deployed applications. Echo’s funding release cites its own research claiming that official Docker images can contain well over 1,000 vulnerabilities. That number is not a universal benchmark; results depend on the image tag, package set, scanner, vulnerability database and scan date.

Echo’s model: replace the image, not just the finding

Traditional scanners begin with an existing image and match installed packages against vulnerability databases. They provide visibility and may recommend upgrades, policy changes or runtime controls, but they do not necessarily remove the vulnerable component.

Echo says it starts with a controlled build process, rebuilds images from source, keeps only required components, hardens the result and continuously rebuilds or patches it. Its product page says releases include signed attestations, SBOMs, provenance and VEX data, with build infrastructure claimed to meet SLSA Level 3. These are vendor-stated capabilities; the available coverage does not independently audit Echo’s implementation or attestations.

Echo also advertises an enterprise remediation commitment: critical and high-severity CVEs are triaged within 24 hours and fixed within seven days. Buyers should confirm which artifacts, severities, exclusions, support channels and clock-start conditions the commitment covers.

What the autonomous agents do

Echo and its funding coverage describe agents that automate the maintenance loop:

1. Monitor vulnerability disclosures and other security information.
2. Determine which maintained images and libraries are affected.
3. Research or develop a fix, or select an appropriate update.
4. Apply the change and rebuild the artifact.
5. Run compatibility tests.
6. Generate a pull request for human review.

Echo said a team of about 35 people maintained more than 600 secure images in December 2025, a scale it argues would otherwise require hundreds of security engineers. That is a company claim. “Autonomous” also does not establish that agents can publish directly to production. A prospective customer should ask which changes require approval, how regressions are detected, what happens when no upstream patch exists, and how the build process prevents a malicious or incorrect agent change.

Does changing one Dockerfile line really make migration easy?

Echo markets a drop-in migration: replace the upstream image reference in a Dockerfile with the corresponding Echo image.

# Before
FROM python:3.12

# Echo-style replacement (exact registry path varies by account)
FROM <Echo-registry>/<corresponding-python-image>:3.12

The syntax is simple, but behavioral equivalence is workload-specific. Before production adoption, test:

• shells, entrypoints, default commands and health probes;
• glibc versus musl behavior and dynamic linker expectations;
• CA certificates, timezone data, locales and filesystem paths;
• users, groups, permissions and numeric IDs;
• native extensions and build-stage dependencies;
• architecture support and pinned digests;
• debugging tools, package-manager assumptions and sidecars.

Echo says its image variants and AI lab are designed for different development and production needs and compatibility testing. That remains a vendor assertion, not a substitute for testing the customer’s own workloads.

What “CVE-free” does—and does not—mean

“Zero CVEs” should be read as zero known, scanner-detected vulnerabilities in a specified artifact at a specified time. CVE databases are not instantaneous or complete, scanners disagree, and a vulnerability may be undisclosed or not yet assigned a CVE.

• Application code can remain vulnerable even when the base image is clean.
• Misconfiguration, excessive privileges, exposed services, leaked secrets and insecure Kubernetes settings are separate risks.
• Severity depends on exploitability and deployment context.
• Removing utilities can reduce attack surface but make incident response and debugging harder.
• Version pinning improves repeatability but can delay updates if maintenance is neglected.
• A rebuilt binary can behave differently despite compatibility tests.

For any numerical “zero” claim, request the image digest, scanner, database version, scan date and policy used.

Compliance and enterprise controls

Echo markets FIPS-validated cryptographic modules, STIG-hardened variants, SPDX and CycloneDX SBOMs, signed attestations, provenance, VEX data, audit support and POA&M assistance. Its FedRAMP-oriented materials may help an organization assemble evidence, but adopting Echo does not make a system FedRAMP-authorized or automatically compliant. Responsibility remains with the system owner and authorization boundary.

Echo also positions its artifacts for the EU Cyber Resilience Act, NIS2 and DORA. Applicability and deadlines vary by product and jurisdiction; these materials should be treated as compliance-support features rather than guarantees.

Integrations and catalog claims

Echo lists integrations with Docker, GitHub Packages, Harbor, Nexus, Red Hat Quay, JFrog, Google Artifact Registry and other registries on its integrations page. The December 2025 announcement cited more than 600 secure images, while the current website describes thousands of secure artifacts. The sources do not define those catalog scopes well enough to treat the figures as directly comparable.

How Echo compares with alternatives

These categories are not interchangeable. A scanner can remain useful alongside Echo because artifact hygiene does not replace runtime detection, policy enforcement or application testing.

When Echo may justify an enterprise purchase

Likely strong fit

• A large container estate has a persistent inherited-CVE backlog.
• Security engineers spend substantial time triaging base-image findings.
• Customers or regulators require FIPS, STIG, SBOM or provenance evidence.
• Platform teams want to preserve familiar image families and CI/CD workflows.
• A contractual remediation SLA has material operational value.

Potentially poor fit

• The organization already operates a mature, well-funded hardened-image pipeline.
• Workloads require distributions or packages outside Echo’s catalog.
• The main risks are proprietary code, identity, APIs or runtime configuration.
• The organization cannot accept dependence on a third-party image builder.
• The team needs runtime detection more than cleaner artifacts.

Questions to put in a proof of concept

• Does the exact runtime, OS family, version and architecture exist?
• Are build-stage and runtime-stage images covered?
• Can signatures, provenance, SBOMs and VEX records be verified independently?
• Are private registries, disconnected environments, rollback and emergency exceptions supported?
• What happens if a package cannot be safely rebuilt?
• What does an “artifact” mean for pricing, retention and support?
• What are the SLA exclusions, FIPS certificate mappings and exit or caching rights?

Bottom line

Echo’s meaningful differentiator is not simply that it uses AI. It is the combination of a managed secure-image catalog, automated maintenance and a claimed one-line migration path. That could be valuable for enterprises with many similar workloads, recurring inherited-image findings or regulated evidence requirements.

The trade-off is a shift in supply-chain trust—from arbitrary upstream contents to Echo’s build infrastructure, source-selection logic, signing keys, agent workflows and commercial continuity. Treat “CVE-free” as a scoped reduction in known image vulnerabilities, verify the claims with your own scanner and workload tests, and compare the total cost with a hardened-image program you can operate and reproduce internally.

Quick Recap

Bestseller No. 1

Nimo AI NAS, Agentic Computer Mini PC and AI Server, AMD Ryzen 7 PRO 8845HS(up to 5.1 GHZ, beat i5-1235u) up to 132TB ZFS Hybrid Storage, Dual 10GbE for 24hr AI Agent

$1,999.99

Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.