Intro
Beluga is building a trusted hub where users can learn about and engage with leading crypto ecosystems. A core part of that experience are Starter Packs: guided onboarding flows that educate users and drive on-chain engagement through rewards.
As Beluga expanded Starter Packs to support richer actions - like token swaps and staking - the existing implementation couldn’t keep up. They partnered with Hypotenuse Labs to rebuild the Starter Pack development pipeline for speed, consistency, and reliability.
Summary
Making Crypto Onboarding Scalable and Engaging
Hypotenuse Labs rebuilt Beluga's Starter Pack system into a modular, self-serve platform - reducing development time per Starter Pack by 80%+ (from approximately 1 month to under 1 week), improving test coverage for on-chain reward flows, and enabling new engagement features such as staking and swaps.
Beluga shipped campaigns more predictably with fewer launch issues, grew its user base 8x, and increased partner confidence in the platform.
We introduced a self-serve JSON templating system, improved automated testing coverage for on-chain logic, and supported new crypto features like staking via ERC-4626 and ERC-7540.
The Challenge
Beluga's Starter Packs were originally built as bespoke, pack-by-pack implementations. Over time, this created duplication, inconsistent quality, and a growing surface area for bugs - especially as on-chain actions became more complex.
Key constraints:
Slow delivery: each new Starter Pack took ~1 month, with engineering as the critical bottleneck.
Low reuse: UI and backend logic were repeatedly reimplemented, compounding tech debt.
Fragile reliability: limited end-to-end coverage for on-chain reward flows increased the risk of production regressions and delayed launches.
Our Solution + Process
A modular Starter Pack framework
We partnered with Beluga’s design team to create a library of reusable templates and components that can be composed to build new Starter Packs without reinventing the UI and flow logic each time.
Self-serve configuration via JSON templates
We introduced a JSON-driven templating system that allowed non-engineering stakeholders to assemble and iterate on Starter Packs safely within defined constraints - while still producing consistent, production-grade experiences.
To prevent config errors from reaching production, we implemented schema validation and guardrails so malformed packs could be caught early (before deployment).
Automated testing for backend + on-chain reward flows
We built a test automation suite that covered:
new and existing backend endpoints,
reward eligibility logic,
end-to-end on-chain flow regressions.
This reduced manual regression cycles and improved confidence when shipping new pack types.
New on-chain engagement modules
To expand Starter Pack capabilities and increase retention, we added support for:
staking flows via vault-style contracts (e.g., ERC-4626),
token swaps,
user vault interactions and reward-triggering actions.
The Results
Shipping velocity and consistency: 80%+ faster pack development: from approximately 1 month to under 1 week
Engineering no longer blocked launches: partnerships and marketing could drive timelines
Fewer surprise launch bugs: improved validation + automated coverage reduced regressions
Growth and trust: 8x user growth: from approximately 10,000 to 80,000
Higher partner confidence: fewer support escalations and smoother campaign launches
Richer engagement loops: staking and ongoing on-chain actions aligned incentives with long-term usage