As an engineering team grows past a handful of people, developers start spending more time waiting on tickets — "can I get a staging database," "can someone provision this service" — than actually writing code. Platform engineering is the practice of building a self-service layer so developers can get what they need themselves, while the platform team keeps guardrails like security, cost, and compliance enforced underneath. The tools below (sometimes called "internal developer platforms" or IDPs) are how teams build that layer without starting from scratch.
The short version: for a fast, ready-made self-service portal, Port and Cycloid get you running in about a week. For governance and scorecards at a larger org, Cortex and Humanitec focus more on enforcing standards than DIY speed. If you'd rather own the whole stack and avoid vendor lock-in, Kratix and Rivora are free and open source. And if your actual pain point is ungoverned Terraform rather than a missing developer portal, env0 solves that narrower problem directly.
Port
Port is a self-service software catalog with golden-path automation and scorecards, pitched as faster to stand up than building your own on Backstage — typically about a week instead of months. It's free for up to 15 seats and 10,000 entities, then $30-40/seat/month as you add SSO and scale to hundreds of thousands of entities; enterprise pricing is custom.
Good for: platform and DevOps/SRE teams that want self-service infrastructure without dedicating a large team to build it. Watch out for: per-seat pricing adds up quickly past a few dozen engineers, and it overlaps heavily with Cortex — the choice between the two is more about emphasis than raw features.
Cortex
Cortex automatically maps your services into a catalog and uses scorecards to track production readiness, compliance, and now AI-coding-agent adoption. There's no self-serve signup — pricing is custom and requires booking a demo.
Good for: engineering leaders and platform/SRE teams at larger organizations who need service-quality visibility across many teams, backed by SOC 2 Type 2 and ISO 27001 certification. Watch out for: no public pricing at all, and it leans more toward measurement than self-service automation compared to Port.
Humanitec
Humanitec is an orchestration layer that lets developers request cloud resources themselves instead of waiting on a ticket queue, while ops keeps security, cost, and compliance guardrails enforced automatically. It's positioned to integrate with what you already have rather than requiring a migration.
Good for: platform engineering and DevOps teams automating governed cloud provisioning. Watch out for: pricing starts around $2,199/month for 5 users, with enterprise and self-hosted only available on quote — a real commitment for a small team.
Cycloid
Cycloid gives platform teams a ready-made self-service portal — developers fill out simple forms to spin up infrastructure instead of filing a ticket — built GitOps-first with infrastructure-as-code baked in, plus tracking of both cloud cost and carbon footprint. It was recognized in the Gartner Hype Cycle 2026 for Platform Engineering.
Good for: platform, DevOps, MSP, and public-sector teams managing multi-cloud who want a portal ready to go rather than building on Backstage themselves. Watch out for: pricing isn't published anywhere, so budgeting means a sales conversation, and the market here (Port, Humanitec, Cortex) is crowded.
Kratix
Kratix is an open-source toolkit for a platform team to build a self-service capability once — "spin up a database," for example — and expose it consistently through an API, CLI, or portal to developers, automation scripts, and now AI agents via MCP support.
Good for: platform teams that want to build their own internal developer platform with zero vendor lock-in and no license cost self-hosted. Watch out for: it demands real implementation effort compared to a ready-made SaaS portal, and enterprise/agentic tiers aren't publicly priced.
env0
env0 doesn't replace your infrastructure-as-code setup — it's a governance layer that runs your existing Terraform, OpenTofu, Pulumi, or CloudFormation through a pipeline with approvals, drift detection, and compliance policies, instead of engineers running terraform apply from their laptops. Pricing starts around $1,500/month with unlimited users included.
Good for: platform, DevOps, and infrastructure teams governing IaC at scale who don't need a full developer portal — just control over what's already there. Watch out for: the entry price is steep for a small team, and higher tiers are quote-only.
Rivora
Rivora is an open-source "engineering memory" tool that pulls GitHub, CI logs, Kubernetes, and Sentry into one place, so a team investigating an incident isn't stuck re-piecing together the same story from five dashboards every time. It's free under Apache-2.0 and written in Rust as a lightweight, self-hostable CLI.
Good for: engineering and platform teams who want a shared, persistent memory across GitHub, CI/CD, Kubernetes, and observability tools, and are comfortable running something young. Watch out for: it's a very early project — no public traction or case studies yet, and far fewer integrations than an established platform like Cortex.
All seven, side by side
| Tool | Best for | Pricing | Open source |
|---|---|---|---|
| Port | Fast self-service portal, mid-size teams | Free to 15 seats, then $30-40/seat/mo | No |
| Cortex | Governance & scorecards at large orgs | Custom, sales only | No |
| Humanitec | Guardrails bolted onto existing workflows | From ~$2,199/mo (5 users) | No |
| Cycloid | Self-service + cost/carbon tracking | Non-public | No |
| Kratix | Build-your-own platform, AI-agent ready | Free self-hosted | Yes |
| env0 | Governing existing Terraform/IaC | From $1,500/mo | No |
| Rivora | Cross-tool incident/engineering memory | Free | Yes |
None of these tools is a universal answer — a five-person startup and a 500-engineer enterprise have almost nothing in common when it comes to platform needs. If you're small and just want to stop the ticket-queue bottleneck fastest, start with Port or Cycloid. If you're large and need standards enforced across dozens of teams, look at Cortex or Humanitec. If engineering time is cheaper than license fees for you, Kratix and Rivora let you build exactly what you need for free. And if the real problem is just ungoverned Terraform, don't buy a whole platform — env0 alone might fix it.