Senior Full-Stack & Mobile Developer (Tech Lead)

Tech Lead (Senior Full Stack & Mobile)

About SEMA Rewards

SEMA Rewards is a clipping platform live in production in Indonesia. Brands and agencies fund campaigns. Clippers turn the brand's content into short videos, post them on their own accounts across all major social platforms, and get paid real money for the views they bring in. The platform handles everything in between: campaign discovery, submissions, view tracking, reviews, earnings, wallets and payouts.

We're a small team and we ship every day.

The role

This is a lead role reporting directly to the founder. You'll own the whole technical side of the product: the web app, the backend, the Android and iOS apps, our video tools, the infrastructure and the releases. You and the founder decide together what to build next.

We're not looking for someone who waits for instructions. We want someone who talks to clippers and brands regularly, learns where they struggle, and turns that into improvements, week after week.

What you'll do

1. Keep shipping. Build features, fix bugs, and release to staging and production on a steady rhythm.

2. Keep the money right. Wallets, earnings ledgers, payout approvals, campaign funding and payment gateway integrations (payments in, disbursements out, webhook handling) must be correct, auditable and fast.

3. Own the mobile apps. Build our Android and iOS apps in React Native, sharing code with the web app. Handle store submissions and releases on both stores, push notifications and over the air updates.

4. Harden our integrations. Our third party data providers fail often. Retries, timeouts and processing that's safe to replay are daily work here, not edge cases.

5. Stay close to users. Talk to clippers and brands, read support messages and community chats, and watch where people get stuck. Turn that into small releases, then check back on whether the change helped.

6. Simplify our flows. The product grew fast and some paths take too many steps. Start with two: the clipper's path from signup to first payout, and the brand's path from launching a campaign to seeing results.

7. Build our video tools. Give clippers the tools to turn long videos into great short clips.

8. Run production. Own deploys, monitoring and incident response. Watch performance, failed jobs, rate limits, duplicate accounts and fake views. Keep the automated test suite that gates every deploy strong.

9. Measure. Track the funnel so we know where users drop off and whether a change worked.

10. Report daily. At the end of each working day, send the founder a short written update: what shipped, what's next, what's blocked, what you heard from users, and one number that moved.

Your first 90 days

1. Weeks 1 and 2: run the whole stack locally, ship your first fixes, learn how deploys work, and map every user flow.

2. Month 1: own releases, talk to at least 10 clippers and a handful of brands, and deliver a short audit of our flows with the changes you'd make first.

3. Month 3: a shorter path to first payout, a funnel dashboard we trust, a steady feedback loop with users, the mobile apps on a set release schedule, and a first prototype of our video tools.

How we work

1. Nobody hands you a ticket list. You bring proposals backed by what users say and what the data shows, and agree on priorities with the founder.

2. Trunk based development with PRs. Every change is reviewed, gated by CI, and soaks on staging before production.

3. We build with AI coding agents every day. You'll direct them and review their code as strictly as any pull request.

4. You work directly with a non technical founder and decisions happen fast. Honest estimates and specific risk flags matter as much as code.

What you bring

1. 6+ years building production software, including at least 2 years as the lead engineer on a product.

2. Strong TypeScript across the stack. React (ideally NextJS), and clean delivery from designs on web and mobile.

3. Solid backend skills: NodeJS and REST APIs.

4. Real PostgreSQL depth: schema design, safe migrations on a live database, query tuning, connection pooling and its failure modes, lock diagnosis. "Knows SQL" is not this.

5. Financial correctness instincts. You think in transactions, row locks, idempotency and invariants without being asked. You know why a payment webhook must be safe to replay and why a balance check belongs inside the transaction that spends it. You've worked on money flows where a bug costs real money.

6. Production React Native. At least one app shipped to both Google Play and the App Store and maintained after launch. You know app signing, TestFlight, the Play Console and store review. Expo experience preferred.

7. Hands on video experience: processing and transcoding with FFmpeg, and video playback or editing in a web or mobile app.

8. Production operations. You've handled a live outage and know that rollback comes before diagnosis.

9. Defensive integration engineering. You assume external APIs will lie, duplicate or vanish.

10. Initiative and product sense. You find problems yourself by talking to users and reading the data, design a simpler flow, and show with data that it worked.

11. Evidence based communication. You say "I measured X," not "probably Y," and nobody has to chase you for updates.

12. Fluent Bahasa Indonesia and English.

Nice to have

1. Creator economy, social media or marketplace experience

2. Video rendering or editing on the GPU (WebCodecs, WebGL or WebGPU, AVFoundation, MediaCodec)

3. AI video features: automatic captions, highlight detection, reframing to vertical (Python, speech to text, computer vision, LLMs)

4. Running user interviews or usability tests

5. Fraud prevention: duplicate accounts, fake engagement

6. Building for midrange Android phones and slow connections

7. Cloud infrastructure and CI/CD, including GPU workloads

How to apply

Apply through this job post with your CV, your GitHub or portfolio, links to apps you've shipped, and your expected salary range. Add a few sentences about a problem you found yourself by talking to users, before anyone asked you to fix it: how you found it, what you changed, and how you knew it worked.

Our process: a short intro call, a take home exercise, and a final interview with the founder. We'll share our full stack on the first call.

The take home exercise

Two parts: a product review (about 1 hour) and a small app build (about 1 day). Send both within 5 days of the intro call. We'll tell you where to send them on the call.

Part 1: Use our product like a new clipper (about 1 hour)

Go to our website and use it the way a new clipper would: sign up, look around, find a campaign, and get as far toward submitting a clip as you can. Then write up to one page:

1. The three biggest problems you found. For each: what happened, and why it matters to a clipper.

2. How you'd confirm them. What you'd ask real clippers, or which numbers you'd look at, to check that each problem is real and common.

3. What you'd ship first, and why.

We're looking for problems that stop people from earning, not typos.

Part 2: Build a small AI clip app (about 1 day)

Build a small React Native app (Expo is fine) that finds the best moment in a long landscape video and turns it into a short vertical clip:

1. Pick a video from the phone. Any landscape video a few minutes long works.

2. Let AI pick the moment. Use AI to analyze the video and suggest which part is most worth clipping, for example from the transcript, the audio or what's on screen. Show the user why it picked that part.

3. Let the user adjust. The user can accept the suggestion or change the start and end time. The clip must be 60 seconds or less.

4. Make it vertical (9:16). Crop it, fit it on a background, or anything else; your choice.

5. Add one line of caption text, burned into the video.

6. Export it: save it to the phone or open the share sheet.

Use any AI model, API or library you like. The processing can run on the phone, on a small server you write, or both.

Explain your stack. In your README, list the framework, libraries, AI model or API and video tools you used, where each step runs (phone or server), and why you chose them over the alternatives.

AI coding tools are welcome; tell us which parts they wrote and how you checked them.

Send us:

1. A link to the repository, or a zip

2. A README covering how to run it, your stack and why you chose it, your trade offs and what you'd do next, and how you used AI coding tools

3. A 2 to 3 minute screen recording of the app working

Final interview: you'll walk the founder through your code and make one small change to it live, so choose an approach you can explain.