Infrastructure

Global application hosting for stateful workloads

Homeground · Deploy apps with built-in replicated SQLite, close to your users worldwide

WatchingDraft scoreMature market · since 2007 · AI estimate

52PWAS score

Confidence

Based on 9 of 9 factors

Overview

AIThis description was written by a model from public signals. The facts behind it are listed under Evidence.

Problem: Deploying stateful applications globally requires managing databases, instance placement, and data replication complexity. Serverless platforms excel at stateless workloads but force developers to choose between serverless simplicity and the complexity of managing state across regions, or between stateless convenience and the need to run persistent services close to users.
Solution concept: A distributed platform that abstracts away state management complexity by combining lightweight VM orchestration with built-in replicated SQLite databases, allowing developers to deploy stateful applications globally with local autoscaling and automatic data consistency, without managing infrastructure or external databases.
Who has it: Application developers and small teams building web applications and services that need state management, autoscaling, and global distribution without infrastructure complexity.
How they handle it today: Developers today either use traditional PaaS platforms like Heroku that handle deployment but lack global distribution, use serverless functions (Lambda, etc.) that are stateless and force external database management, or use infrastructure-as-code tools to manage their own VMs and databases across regions.
Why now: Lightweight VM technology and efficient replication (via CRDT and object storage) now make it practical to provide the simplicity of classic PaaS with the global distribution and autoscaling benefits previously only available through complex infrastructure management or serverless with external state.
How it could charge: Subscription pricing based on compute (VM instances), memory usage, storage, and data transfer; or per-region pricing with a base cost for global deployment.

Score breakdown

A model sorted each factor into a rubric bucket; code turned buckets into numbers. Unknown factors are left out of the score and lower confidence instead.

Market validation

Confidence 95 · 4 evidence · Clear demand evidenced by active competitors (Fly.io, Heroku, AWS Lambda) solving similar problems, plus community interest on Hacker News with 141 points and 25 comments.

Revenue evidence

Confidence 95 · 3 evidence · Several established competitors with substantial publicly stated revenue: AWS Lambda has paid plans, Fly.io offers 9 paid plans up to $2,500/month, and Heroku offers 9 paid plans up to $750/month.

Growth

Confidence 95 · 1 evidence · Modest interest shown by Hacker News engagement (141 points), but no public growth metrics for the opportunity itself; the category of global application platforms is established but not showing breakout growth signals.

Problem severity

Confidence 70 · 1 evidence · The problem causes costly friction: developers regularly lose hours managing data consistency and autoscaling complexity across regions, forcing difficult tradeoffs between simplicity and functionality.

Economic proximity

Confidence 30 · Judged from the description · The product directly controls compute resource consumption and charges for it; pricing models (per-instance, memory, storage, data transfer, per-region) sit at the point of money changing hands in infrastructure spending.

Distribution

Confidence 30 · Judged from the description · A platform hosting service is shared internally within teams and organizations, but deployed applications themselves are seen by end-users; the platform itself operates as a shared infrastructure tool with indirect distribution through hosted products.

Buildability

Confidence 70 · 1 evidence · Building requires multiple complex integrations (lightweight VM orchestration, replicated SQLite databases, data consistency protocols, global networking) and heavy infrastructure requirements, though modern VM technology enables millisecond startup times.

Differentiation

Confidence 70 · 2 evidence · Clear gap between serverless (stateless) and traditional infrastructure: developers explicitly want stateful apps without complexity, and incumbents leave this tradeoff open (AWS Lambda is stateless, Fly.io requires more infrastructure knowledge than Heroku).

Retention and timing

Confidence 30 · Judged from the description · Application hosting is a recurring workflow (developers deploy and iterate continuously), and this represents a clear timing tailwind as lightweight VM technology and developer preference for simplicity have matured.

Risk penalties

Extreme competitionThe space is occupied by entrenched, well-funded competitors: AWS Lambda dominates serverless, Fly.io is an established venture-backed platform with 4 years of engineering, and Heroku is a market-proven classic PaaS with significant infrastructure investment.−10

Score history

52 · confidence 65 · config v1First score · 15 Oct 2026

Evidence · 9 items

Market validation

Hacker News post received 141 points and 25 comments
Verified
Hacker NewsRetrieved 15 Oct 2026
The author spent 4 years building Fly.io and saw limitations in the serverless/stateful app tradeoff
Public claim
I helped build Fly.io for 4 years and shared enthusiasm for the founders' vision of a 'global Heroku'
Hacker NewsRetrieved 15 Oct 2026

Revenue evidence

AWS Lambda's pricing page lists 1 paid plan, at $0.20 by usage, plus a free plan
Verified
Request charges : $0.20 per million requests
aws.amazon.comRetrieved 11 Oct 2026
Fly.io's pricing page lists 9 paid plans, from $29 to $2,500 a month, plus prices on request for larger customers
Verified
Standard $29 / mo · Premium $199 / mo · Enterprise From $2,500 / mo · HIPAA package at $99/mo · Plan CPU and RAM Per month Basic shared-2x · 1GB $38 · Starter shared-2x · 2GB $72 · Launch performance-2x · 8GB $282 · Scale performance-4x · 32GB $962 · Performance performance-8x · 64GB $1,922 · Custom plan Committed spend, volume discounts, invoicing and dedicated capacity. Tell us what you are running. Contact us
fly.ioRetrieved 15 Oct 2026
Heroku's pricing page lists 9 paid plans, from $5 to $750 a month
Verified
Eco $5 0.5 GB · Basic $7 0.5 GB · Standard-1X $25 0.5 GB · Standard-2X $50 1 GB · Performance-M $250 2.5 GB · Performance-L $500 14 GB · Performance-XL $750 62 GB · Private-S $125 1 GB · Private-M $250 2.5 GB
heroku.comRetrieved 9 Oct 2026

Problem severity

Autoscaling across regions requires managing data consistency, which existing solutions force developers to micro-manage
Public claim
This lets Kedge autoscale services close to your users while each instance queries its local replica for eventually-consistent data, with no need to micro-manage instance or volume placement
Hacker NewsRetrieved 15 Oct 2026

Buildability

Lightweight VMs can now create instances in milliseconds, enabling practical serverless-like performance with state support
Public claim
Kedge has a fast VM orchestrator that can create code sandboxes or scale service instances in 3ms, using a combination of forkable VM snapshots and a tree of warm pools
Hacker NewsRetrieved 15 Oct 2026

Differentiation

Developers want to build stateful applications (server-rendered HTML apps with forms, votes, and auth) without state management complexity
AI inference
The model stated this without a quote found in the source.
Hacker NewsRetrieved 15 Oct 2026

Retention and timing

The platform is in early preview with no billing yet, indicating early monetization stage
Public claim
There's no billing yet, so the pricing page is an estimate
Hacker NewsRetrieved 15 Oct 2026

Global application hosting for stateful workloads

Overview

Build this business

Verdict

Score breakdown

Risk penalties

Score history

Validation gate

Evidence · 9 items

Market validation

Revenue evidence

Problem severity

Buildability

Differentiation

Retention and timing

Market

Existing products

Build

Legal reconstruction check

Community