Curated Developer Resources

Our own pick for this site, and the easiest way to get a Next.js app live. Git-push deploys, preview URLs per PR, and a genuinely usable free hobby tier. The tradeoff: serverless function pricing can climb fast once a project has real backend traffic, so it suits frontend-heavy apps better than compute-heavy ones.

Closer to a full PaaS than Vercel — good when you need a long-running backend, a worker process, or a database sitting next to your app without stitching together separate services. Usage-based pricing means costs scale with the app rather than a flat tier, which is easier to reason about once you're past hobby scale.

A solid, more affordable landing spot if you're moving off Heroku — static sites are free and web services start around $7/mo. Cold starts on the free web-service tier are noticeable, so it's a better fit for background services and internal tools than latency-sensitive public APIs.

Worth reaching for specifically when latency to users in multiple regions matters — it runs your actual Docker container close to users worldwide, rather than routing everything through one region. More operational surface area than Vercel or Render (you're managing containers, not just pushing code), so it pays off more once you actually need that geographic spread.

The pick if you want Postgres plus auth, storage, and realtime subscriptions bundled together without wiring each piece up separately — genuinely useful for shipping an MVP fast. It's a real Postgres database underneath (not a proprietary layer), so you can always outgrow the bundled parts and connect directly with standard tooling later.

Distinct from the others here for one reason: branching your database schema the same way you branch code, with zero-downtime migrations built around that workflow. Worth it specifically if your team ships frequent schema changes and has been burned by risky migrations before — less of a differentiator for a project with a stable, rarely-changing schema.

Serverless Postgres with the same branching idea as PlanetScale, but for Postgres instead of MySQL — a natural fit if you're already committed to Postgres and deploying on Vercel, since scale-to-zero pricing matches serverless traffic patterns well. Expect a cold-start delay on the free tier after idle periods.

The one to reach for when you need Redis or Kafka but don't want to pay for an always-on instance sitting idle — it's priced per request with no idle cost, which fits serverless functions (Vercel, Lambda) far better than a traditional always-connected Redis client model.

The default choice for error tracking for a reason — stack traces arrive with the actual source context, breadcrumbs leading up to the error, and enough free-tier volume to cover a small project without paying. It's specifically for catching and diagnosing exceptions, not a general analytics tool — pair it with something like PostHog below if you also need product usage data.

Covers the product-analytics side Sentry doesn't — funnels, session replays, feature flags — and being open-source means self-hosting is a real option if sending user session data to a third party is a concern. The cloud free tier is generous enough for early-stage products; self-hosting only really pays off once volume or data-residency requirements justify the operational overhead.

Answers a different question than Sentry or PostHog: not "did an error happen" but "is the thing actually up right now," checked proactively from the outside rather than waiting for a user-triggered error to report in. Worth adding once you have an API or endpoint where downtime itself (not just bugs) needs to page someone.

Pre-built, production-quality component markup from the team that maintains Tailwind itself — the fastest route from a blank page to a polished layout if you're already using Tailwind. It's a one-time paid purchase, not free like shadcn/ui below, and you're copying static markup rather than an installable component, so plan for some manual upkeep as your design evolves.

The free counterpart to Tailwind UI, but structurally different in a way that matters: components are copied directly into your codebase (not installed as a dependency), built on accessible Radix primitives. That means you can freely restyle or modify a component without fighting an npm package's API — the cost is that updates to the library don't reach your project automatically.

A clean, consistent open-source icon set that's become the default pairing for shadcn/ui and similar Tailwind-based stacks — over 1,000 icons with matching stroke weight and style, so a UI doesn't end up with visibly mismatched icons pulled from different sets.

A terminal rebuilt around blocks of output you can select, share, and search — rather than a raw scrolling text stream — plus AI-assisted command autocomplete. The learning curve is small since it's still a real shell underneath, but it's a genuine workflow shift if you've used the same terminal muscle memory for years.

Built specifically for engineering teams rather than adapted from general project management — keyboard-first navigation, fast issue creation, and Git branch/PR integration that ties commits back to issues automatically. It intentionally does less than heavier tools like Jira; that constraint is the point, not a missing feature, for teams that found Jira's flexibility slowed them down.

A lighter-weight alternative to Postman for testing APIs, with a command-line client alongside the desktop/browser app — useful if you want to script and version-control your API requests rather than click through a GUI collection. Postman still has the edge for large teams needing shared collections and mock servers; HTTPie wins for quick, individual iteration.

.com registration from around $9/year with free WHOIS privacy protection included by default — many registrars charge extra for that separately. Not a hosting or DNS-performance play; it's specifically a straightforward, low-cost place to register and manage domains.

A transactional email API built around React Email components, so your email templates are written and reviewed the same way as the rest of your frontend instead of hand-written HTML strings. The free tier (3K emails/mo) comfortably covers a small app's signup and notification emails; it's built for transactional mail specifically, not bulk newsletter sending — that's what Beehiiv below is for.

Purpose-built for newsletter publishing rather than adapted from a general email-marketing tool — built-in referral programs and monetization options that most transactional email APIs simply don't offer. The right layer above Resend: Resend for "your app needs to send an email," Beehiiv for "you're running an actual newsletter publication."