Cuesoft Developer Labs
Ship real software before you graduate.
An open-source internship for students and researchers. You work on production codebases used by real people — Apparule, Expendit and Upstat, reviewed by the engineers who maintain them, and you work the way they do: AI-native, with coding agents in the loop.
Unpaid · data provided · remote · 3 or 6 months
Quarterly intakes
Don't Feel Like Scrolling? Just Ask.
Ace knows every role, project and term on this page. Interview us before we interview you.
Ace · the CueLABS™ assistant
Ace · the CueLABS™ assistantAce can make mistakes — the apply page is the source of truth.
Apply now →A working department, not a course.
CueLABS™ is Cuesoft’s open-source project department. Interns join a squad, pick up real issues, and ship to codebases that are live and public. Nothing here is a sandbox exercise. The squads work AI-natively, coding agents and automation in the loop, because that is how software is built now.
- Duration
3 or 6 months
Long enough to own something end to end. Three months suits a single industrial-training placement; six suits research work.
- Location
Remote
The team works across Nigeria and beyond. Standups, reviews and pairing happen online, so where you study does not decide whether you can take part.
- Compensation
Unpaid · data provided
This is not a paid role. Because it is remote, interns are afforded data so connectivity is never the reason someone drops out.
- Output
Public commits
Everything is open-source under MIT. Your pull requests carry your name on GitHub for good — a portfolio you keep after you leave.
Built for people who need the experience, not the certificate.
Undergraduates on industrial training
Computer science and engineering students doing SIWES or an IT placement. CueLABS™ counts as the real thing: a supervised placement with a working team, a codebase, and evidence you can hand your department.
- Placement documentation your school can verify
- A mentor who reviews your code, not just your attendance
- Public commits you can show at your defence
Graduate research students
Masters and PhD students working on machine learning, fintech or reliability. Our projects carry open problems with real data behind them, and the results stay publishable because the code is MIT.
- Open problems in anomaly detection, ledgers and SLOs
- Production telemetry and transaction data to work against
- Nothing locked behind an NDA — publish what you find
Eight ways in. Pick the one you want to be good at.
Engineering & design
Frontend Developer
Next.js · React · TypeScript
Product surfaces and dashboards across all three codebases.
Backend Developer
Go · gRPC · Postgres · Mongo
APIs, data pipelines and the services behind every feature.
Mobile App Developer
Flutter · Dart
Apparule’s capture flow, where the camera work actually happens.
UI/UX Designer
Figma · design tokens
Flows, states and the component libraries engineers build against.
DevOps Developer
Docker · Helm · CI/CD · OpenTelemetry
Pipelines, deploys and the observability that keeps them honest.
Research tracks — open to graduate students
ML Research
Anomaly detection · CV
Transaction anomaly models and pose quality control.
Fintech Research
Ledgers · tax logic
Categorisation, financial ratios and filing rules.
Reliability Research
SLOs · telemetry
Failure prediction and reliability signals in Upstat.
Three products in the wild. You work on these.
Not practice repositories. Each one is live, MIT licensed, and open to pull requests today.
Apparule
Two photos. A perfect fit.
The problem
Tailors and fashion SMEs still keep measurements on paper, and shoppers guess at size charts that do not match their body. Both sides lose to returns and remakes.
What it does
Apparule turns two phone photos into a complete, private body-measurement profile, then carries those measurements into every commission.
- Flutter
- Dart
- Go
- Python
- Node.js
- Next.js
Problems worth your term
Pose quality control on noisy phone cameras
Turning landmarks into garment-grade measurements
Escrow-protected commissions
Open issues are labelled and scoped. Your first one is chosen to be small enough to merge in week one.
The stack you will actually touch
Not a syllabus — this is what is running in the three repositories today. You will not use all of it; you will get good at some of it.
- Go
- Python
- Node.js
- TypeScript
- React
- Next.js
- Flutter
- Dart
- PostgreSQL
- MongoDB
- Redis
- ClickHouse
- OpenTelemetry
- Docker
- Kubernetes
- Helm
- GitHub Actions
- Figma
- Git
How it works
- 01
Apply
One form: who you are, the role you want, and how long you can commit. No cover letter, no essay.
- 02
Screening call
A short conversation about what you have built and where you want to get better. We are looking for people who finish things.
- 03
Onboarding week
You get repository access, a mentor, and a first issue chosen to be small enough to land in week one.
- 04
Ship in the open
You work the board like any other engineer — branch, pull request, review, merge, AI tooling in the loop. Your name is on every commit, permanently.
- 05
Reference and portfolio
You leave with merged code in a public repository, a documented placement, and a reference from the engineer who reviewed your work.
You will learn to contribute the way real projects expect.
Every CueLABS™ repository follows one published engineering standard — the same rules, templates and CI gates in each project. It is open source itself, so you can read exactly what will be expected of you before you apply.
- Fork, branch, pull request
- Feature branches off a freshly fetched main. No pushing to main, no self-merging.
- Conventional Commits
- feat: fix: chore: docs: refactor: test: — a readable history you can point an employer at.
- Review before merge
- At least one approving review from a CODEOWNER. You will get real feedback and be expected to answer it.
- Templates for everything
- Issue and pull-request templates that make you state the change, the affected area and how you tested it.
- Community health files
- README, CONTRIBUTING, CODE_OF_CONDUCT, SECURITY and CODEOWNERS — identical across every repository.
- Tested and typed
- Lint, typecheck and tests run in CI on every pull request, so review is about design, not formatting.
This is a pathway, not a dead end.
CueLABS™ sits in the middle of how Cuesoft brings people through: learn the craft, build in the open, then go forward to CueHIRE™ on the strength of what you actually shipped.
- 01
CueTA™
learn from the people who build AI
Structured training if you are starting from scratch or need the fundamentals first.
Visit CueTA™ → - 02You are here
CueLABS™
build real products in the open
You are here. Ship production software on live open-source projects, reviewed by the engineers who maintain them.
You are here. - 03
CueHIRE™
where the work speaks for you
Cuesoft’s talent arm. Interns who do the work are put forward to CueHIRE with a real commit history behind their name — not a certificate.
Visit CueHIRE™ →
What we expect from every intern — be a Cuesoft ACE
- Accountability
- Your name is on every commit, permanently — you own what you ship.
- Collaboration
- Squads, reviews and public boards — nothing here ships alone.
- Excellence
- One published engineering standard, enforced in CI on every pull request.
The full formulation is Cuesoft's — create IMPACT by being a Cuesoft ACE, and it is where Ace, the assistant above, gets its name.
The work happens in the open, in Discord.
Standups, code review chatter, design critique and the occasional argument about naming. You do not need to be accepted to join — come and see how the team works before you apply.
Join the Discord