Platform engineering teams are expected to build IDPs (internal developer platforms) from scratch, combining Kubernetes, Terraform, CI/CD, and observability into cohesive self-service tooling — a massive, ambiguous undertaking that often fails.
Opinionated, deployable IDP starter kits with pre-wired integrations (K8s + Terraform modules + CLI scaffolding + developer portal) that teams customize rather than build from zero. Includes golden path templates for common app types.
This is a top-3 pain point in DevOps/platform engineering right now. Teams are told 'build an IDP' and given a blank canvas. Average IDP build takes 12-18 months and many fail outright. The Reddit thread confirms: platform teams are drowning in complexity, struggling to make Kubernetes, Terraform, CI/CD, and observability work together coherently. The pain is acute, widespread, and largely unaddressed by existing tools which only solve fragments.
TAM is significant but bounded. Target is platform engineering teams at companies with 100+ developers. Estimated 15,000-25,000 such companies globally. At $20k-50k/year average contract, that's a $300M-$1.25B addressable market. Not venture-scale massive, but excellent for a bootstrapped or seed-stage company. The constraint: this is a sophisticated B2B buyer that requires trust, not a mass-market play.
Platform engineering teams at 100+ dev companies already have budgets. They're spending $100k-500k+ in engineer-time building IDPs from scratch. A $20-50k/year product that saves 6+ months of senior engineer time is an easy ROI story. Humanitec, Port, and Backstage-managed providers (Roadie) all prove enterprise willingness to pay in this space. The key is demonstrating time-to-value — 'deploy in days, not months' is a compelling pitch when the alternative is burning a senior team for a year.
This is the critical weakness. A solo dev CANNOT build a credible MVP in 4-8 weeks. The product requires deep expertise across Kubernetes, Terraform, CI/CD (multiple systems), observability, and developer portals — plus maintaining compatibility across cloud providers and versions. Even a focused MVP (e.g., one cloud provider, one CI/CD system, one app type) requires production-grade Terraform modules, Helm charts, working CI pipelines, a developer CLI, and documentation. That's 3-6 months minimum for an experienced platform engineer working full-time. Testing across real environments adds more time. This is fundamentally an infrastructure product — it must work reliably in production or it's worthless.
The gap is real and well-defined. Every existing tool is a building block, not a solution. Backstage = portal layer only. Crossplane = infra provisioning only. Humanitec = orchestration layer only. Nobody ships 'here's a working IDP, deploy it to your cluster, customize it.' The closest analogy is how Rails/Django gave you a working web app to customize vs. assembling Express+ORM+Auth+Templates yourself. Platform teams desperately want this but nobody provides it yet.
Strong subscription fit. Once deployed, the IDP is foundational infrastructure that needs ongoing updates (new K8s versions, Terraform provider updates, security patches, new golden path templates). Teams won't rip it out. Expansion revenue is natural: more developers onboarded = higher tier. Enterprise support tier for production issues is standard. Template marketplace for new app types is a growth vector.
- +Massive, validated pain point — platform teams are drowning and the 'build from scratch' approach has a high failure rate
- +Clear gap in market — nobody ships a complete, deployable IDP; every tool is a partial solution requiring heavy assembly
- +Strong willingness to pay — enterprise budgets exist and the ROI story (months of senior eng time saved) is compelling
- +Natural expansion and retention — infrastructure products are sticky, and per-developer pricing scales with adoption
- +Timing is excellent — platform engineering is at peak hype but most teams are still in the 'struggling to build' phase
- !Technical complexity is enormous — maintaining production-grade K8s + Terraform + CI/CD + portal integrations across versions and cloud providers requires deep, ongoing investment; this is NOT a weekend project
- !Backstage ecosystem gravity — many teams have already committed to Backstage as their portal layer, so you either integrate with it (adding complexity) or compete with CNCF momentum
- !Enterprise sales cycle — target buyers (platform teams at 100+ dev companies) require trust, POCs, security reviews, and long sales cycles; this doesn't fit a 'launch and get revenue fast' model
- !Opinionated vs. flexible tension — the value is in being opinionated (pre-wired decisions) but every company thinks they're special and will want customizations that break your golden paths
- !Potential for incumbents to ship this — Humanitec, Port, or even a Backstage plugin ecosystem could add starter kit functionality and crush a small player
Open-source developer portal framework with plugin architecture for building internal developer platforms. Provides service catalog, software templates, and TechDocs.
Commercial platform orchestrator that provides a reference architecture for IDPs. Abstracts infrastructure behind 'Score' workload spec and dynamic resource matching.
Internal developer portal with a software catalog, self-service actions, scorecards, and workflow automation. Positions itself as a no-code/low-code portal builder.
Open-source framework for building platforms as a product on Kubernetes. Uses 'Promises'
Open-source Kubernetes-based control plane for infrastructure. Crossplane lets you define cloud resources as Kubernetes CRDs. Upbound provides managed Crossplane with a marketplace of providers.
Pick ONE golden path: AWS EKS + Terraform + GitHub Actions + Backstage portal + Datadog observability. Ship a single deployable kit that gives a platform team a working IDP for 'deploy a containerized web service' in under a day. Include: (1) Terraform modules for EKS cluster + networking + IAM, (2) Helm charts for platform services, (3) Backstage software templates for the golden path, (4) GitHub Actions workflows for CI/CD, (5) A CLI tool to bootstrap and customize. Target: 'Your first golden path is live by Friday.' Do NOT try to support multiple clouds or CI systems in v1.
Open-source the basic single-cloud kit to build trust and community (platform engineers hate closed-source infra) -> Paid tiers for: multi-cloud support, additional golden path templates (ML workloads, data pipelines, serverless), SSO/RBAC, compliance templates (SOC2, HIPAA guardrails), priority support -> Enterprise tier with custom golden paths, dedicated support, and SLA guarantees -> Long-term: template marketplace where community contributes golden paths (take a cut)
6-9 months to first paying customer. 3-4 months to build a credible single-cloud MVP that actually works in production. 1-2 months of design partners / beta testing (critical — you need real platform teams validating this). 1-3 months to close first paid deal through the enterprise buying process. Open-source-first strategy could accelerate community building but delays direct revenue. This is NOT a quick-revenue idea — it requires deep technical investment upfront.
- “kubernetes, terraform, and native cloud apis have become far too much for application teams to grok and you need to build guardrails”
- “Building consumable infra so developers can develop”
- “I've seen some awful platform teams”
- “normalized patterns of infra, security, stability, observability, tooling”