Credits burn fast on hyperscalers
$100k of AWS credit is very soon spent as a "panic" project to migrate to the Hyperscaler before the bills actually start to arrive.
Early-stage SaaS, API and developer-tool teams get infrastructure credits, architecture guidance and predictable production hardware to build on. No long-term lock-in, no contract that punishes you when you grow up.
Hyperscaler credit programs are designed to allow fast growing companies to spend thousands of dollars in infrastructure as their company rapidly expands and then 'burn up' the credits just in time for them to realize that the only way out is to leave a suboptimal architecture behind.
$100k of AWS credit is very soon spent as a "panic" project to migrate to the Hyperscaler before the bills actually start to arrive.
By the time your credits have run out your workload will depend on 20 managed services. And that is going to be a rebuild not a migration.
As managed services are added to existing workloads, the complexity of your control planes, SLAs and your billing dimensions increases. Few startups date realize that the simplification of workloads that these credit programs promise, rarely materializes.
And when those credits run out, you can continue to run your same workloads on your same architecture on our VPS plans that you can sign up for and pay for on a month to month basis at the same price that you were locked into with the credits. No lock-in architecture. No huge, complex stack to try and get out of. No dilution of equity for your company.
These credits are designed to scale with the team and match the stage of the company. As the company grows and the team becomes more proficient the amount of credit provided will also grow. These credits can be used for VPS's, IP addresses, for taking snapshots and for scaling up additional power when needed. The credits also come with technical support for Architecture review and planning, for VPS sizing, for scaling advice and for full infrastructure planning and design for the team.
Multi-tenant platforms (e.g. SaaS companies) with paid customers or paid customers in the pipeline. Production workloads which benefit from fixed per-core latency.
API platforms. Backend infrastructure for developer-facing APIs. Pinned cores on Dedicated CPU plans match the deterministic-latency requirement.
Teams shipping developer tools with a hosted backend (auth, telemetry, package registries, CI runners) where predictable infrastructure cost matters.
Small to medium size LLM hosts and supporting services (vector stores, large embedding caches) for AI inference services running on Dedicated CPU or High Memory hardware.
Agencies building SaaS products. Agencies productising client work as SaaS products. The model fits with the agencies' business model of recurring credit from clients for work. The model also fits with the transition that agencies are going through of selling agency style work as SaaS products.
Short-term experiments, crypto mining, traffic arbitrage, mass-email infrastructure, content piracy or affiliate-spam operations.
The application process involves a short note (200-300 words approx.) for us about the team, product and workload and we do 3 step review (no decks) and you can apply to the program immediately afterwards. Architecture review and credit sizing is done during the technical call and the credits are approved and active straight away after that.
From $18.40/mo on a 2-year term. KVM on AMD EPYC, NVMe RAID10, ECC memory. Your price is fixed for the lifetime of the service.