A CCTV surveillance platform built incrementally. Every lab ships a working increment and the evidence that it works — in the same repository.
Same repository, same domain, same constraints from Lab 1 to Lab 8. Requirements written in Lab 1 are still load-bearing in Lab 8.
Each artifact is the next lab's input: dataset → model → service → pipeline → monitors → agent.
Final review is a walkthrough of a running system, not a folder of disconnected notebooks.
| Lab | Chapter | You build | Proof you submit |
|---|---|---|---|
| 01 | Architecture | Local cloud, repo skeleton, requirements and ADR | make verify green from a clean clone |
| 02 | Data engineering | Frame ingestion, sampling, labelling, dataset versioning | Versioned dataset + quality report |
| 03 | Model development | Detector training, tracking, day/night evaluation | Model + experiment record + slice metrics |
| 04 | Deployment | Batch scorer and real-time inference service | Latency/throughput benchmark under budget |
| 05 | MLOps automation | CI/CD/CT: test, build, retrain, promote | Green pipeline + auto-registered candidate |
| 06 | Monitoring | Drift, alert quality, dashboards, runbooks | Dashboard + one real incident postmortem |
| 07 | LLMOps | Incident summariser and natural-language event query | Eval set + cost and latency guardrails |
| 08 | Scale & responsibility | Multi-camera scale-out, privacy and audit controls | Load test + privacy impact assessment |
make up — start the local cloud and services.
make bootstrap — idempotent resource creation.
make seed — realistic cameras and clips, never hand-typed.
make verify — machine-checkable acceptance.
| Weight | Criterion | What earns full marks |
|---|---|---|
| 40% | Reproducibility | Clean clone to green make verify; no hidden state, no manual console clicks. |
| 20% | Design decisions | ADR states the options, the chosen trade-off, and the consequence the team accepts. |
| 20% | Code and tests | Adapters are small and tested; failure paths return actionable errors. |
| 20% | Reflection | Evidence of what broke, how it was detected, and which control now prevents it. |
One docs/ folder accumulates requirements, ADRs, evaluation reports, runbooks and the privacy assessment. It is the revision material for the final walkthrough.
Ten minutes: show an alert, explain the model behind it, show the dashboard, then show what happens when a camera goes dark.
Before any model: a reproducible local cloud, a repository every later lab extends, and a written decision about what “working” means.
A facilities company runs 40 cameras across three sites. They want person and vehicle events with a 60-second alerting budget — and a defensible privacy position.
No model training in this lab. You are building the ground the model will stand on.
| Tool | Why |
|---|---|
| Docker + Compose v2 | Runs the emulator and, later, your services. |
| Floci | Local AWS on :4566 — no account, no token. |
| AWS CLI v2 | Same commands you would run against real AWS. |
| Python 3.11 + venv | Adapters, seed and verification scripts. |
| make | One interface for the whole lab series. |
# Option A — Floci CLI
floci start
eval $(floci env)
# Option B — Docker Compose
docker compose up -d
export AWS_ENDPOINT_URL=http://localhost:4566
export AWS_DEFAULT_REGION=us-east-1
export AWS_ACCESS_KEY_ID=test
export AWS_SECRET_ACCESS_KEY=test
aws s3 ls # must return, not hang
watchtower/
├── compose.yaml floci :4566
├── Makefile up·bootstrap·seed·verify
├── .env.example endpoint, names
├── pyproject.toml boto3, pytest, ruff
├── infra/
│ ├── bootstrap.sh idempotent resources
│ └── teardown.sh
├── src/watchtower/
│ ├── config.py env-driven settings
│ ├── storage.py S3 put/get/list
│ ├── events.py SQS send/receive
│ └── catalog.py DynamoDB cameras/events
├── scripts/
│ ├── seed_cameras.py 3 cameras · 5 clips
│ └── verify.py acceptance, non-zero on fail
├── tests/
└── docs/
├── requirements.md
└── adr/0001-system-architecture.md
Full brief: labs/lab-01.md
$ make verify
[ok] buckets wt-raw wt-curated wt-models
[ok] tables wt-cameras wt-events (+ camera-time gsi)
[ok] queue wt-frame-jobs (+ dlq, maxReceiveCount 5)
[ok] logs /watchtower/lab1
[ok] s3 round-trip 22 bytes
[ok] sqs round-trip 1 message
[ok] seed 3 cameras, 5 clips
PASS lab-01
Port 4566 busy → stop the other emulator. CLI hits real AWS → export AWS_ENDPOINT_URL. Console dead → Docker socket not mounted (optional).
Next: Lab 02 — Video ingestion and the dataset factory