Candle is a viral consumer app helping Gen Z couples build stronger relationships through daily challenges, fun games, and meaningful questions. We recently passed 250,000 users and we’re scaling fast. Our next goal: 1 million users by the end of 2025.

We’re building the communication platform of the future: starting with couples, then expanding to friends, family, and beyond. As an engineering intern, you’ll work on a product seen by hundreds of thousands of people every day and help us scale the backend systems that power all of it.

Candle is a viral consumer app helping Gen Z couples build stronger relationships through daily challenges, fun games, and meaningful questions. We recently passed 250,000 users and we’re scaling fast. Our next goal: 1 million users by the end of 2025.

We’re building the communication platform of the future: starting with couples, then expanding to friends, family, and beyond. As an engineering intern, you’ll work on a product seen by hundreds of thousands of people every day and help us scale the backend systems that power all of it.

What You’ll Do

• Ship fast: Own and ship features end-to-end, used by thousands within hours.
• Scale infra: Help design and optimize systems that can handle millions of notifications, messages, AI inference at scale, and daily interactions
• Optimize performance: Identify bottlenecks, improve reliability, and make Candle feel seamless at scale.
• Bring perspective: Work directly with ex-Twitter, Apple, and NASA engineers, contributing new ideas and whiteboarding solutions to hard problems.
• User impact: Every line of code directly affects couples and friends relying on Candle to stay close and stay in touch. 
• Sample projects of what you could be working on include: overhauling our entire daily notification jobs into a separate reliable microservice, designing and implementing real-time features like a collaborative canvas or a “walkie talkie” feature, creating a recommendation algorithm to dynamically suggest content for couples to explore (this one would be really useful!), etc.

Technologies We Use:

• React Native (Expo) and SwiftUI for the client facing app (iOS and Android)
• Convex for the real-time backend (written in Typescript), including all background tasks, queues, cron jobs, and REST APIs
• Anthropic (Claude 4 Sonnet), OpenAI (GPT 5), and Gemini models to create background agents for tasks like finding local date ideas. We also use Replicate for some image gen models.
• Observability: PostHog for feature flags, Amplitude for analytics, and Sentry for error reporting
• Clerk for user auth(z/n)
• We strongly encourage all candidates to have their own AI setup of choice when coding (Claude Code, Cursor, etc.) to ship faster. Bonus points if you know how to use Git worktrees to speed up your workflow!

Who We're Looking For:

• You're a student (or recent grad) excited by viral consumer products and fast shipping cycles
• You have strong experience with TypeScript and mobile frameworks like React Native (Expo). Swift is optional but a plus!
• You get excited by both backend optimization and user-facing polish, whether you’re speeding up a backend query or refining a feature’s UX 
• You’re curious, self-driven, and relentlessly resourceful
• Entrepreneurial background, have shipped and published apps before (optional but preferred)
• You want your work to matter, to be visible, felt, and loved by users within hours of writing it

We work hybrid in San Francisco, 3 days a week in person and the remaining remotely. You’ll move fast in this internship, break things, and learn more in a few months than most people do in a year. But if you’re the kind of builder who dreams of taking a product from 250k → 1M users, this will be one of the most exciting chapters of your career.

Technologies We Use:

• React Native (Expo) and Swift UI for the client facing app (iOS and Android)
• Convex for the real-time backend (written in Typescript), including all background tasks, queues, cron jobs, and REST APIs
• Anthropic (Claude 4 Sonnet), OpenAI (GPT 5), and Gemini models to create background agents for tasks like finding local date ideas. We also use Replicate for some image gen models.
• Observability: Posthog for feature flags, Amplitude for analytics, and Sentry for error reporting
• Clerk for user auth(z/n)
• We strongly encourage all candidates to have their own AI setup of choice when coding (Claude Code, Cursor, etc.) to ship faster. Bonus points if you know how to use Git worktrees to speed up your workflow!
• Sample projects of what you could be working on include: overhauling our entire daily notifications setup into a separate reliable microservice, designing and implementing real-time features like a collaborative canvas or a “walkie talkie” feature, creating a recommendation algorithm to dynamically suggest content for couples to explore (this one would be really useful!), etc.