03
CRAFT PD Series April 25, 2026 · 8:30 AM โ€“ 12:00 PM PST Virtual (Zoom) โ€” 100% online, no hardware required during the session

Workshop 3: Programming Edge/IoT Systems with AI

BBC micro:bit in the MakeCode simulator + LLM-assisted coding โ€” build a two-node IoT system without any hardware.

C R A F T
BBC micro:bit (online simulator) Physical Computing IoT / Radio Networking LLM-Assisted Coding
Last updated April 27, 2026

Learning Objectives

  • Explain Edge Computing and IoT as a two-role system: sensing nodes and aggregating/display nodes
  • Program a sensor node (collect + radio.send) in the MakeCode simulator
  • Program an aggregator node (radio.receive + aggregate + display) in the MakeCode simulator
  • Pair two simulated micro:bits on a shared radio group so they talk to each other
  • Use an LLM to write, debug, and extend two-node micro:bit code
  • Optionally port MakeCode to MicroPython (python.microbit.org) โ€” simulator or real hardware
  • Align a sensor-and-aggregator activity to NGSS and CTE pathways
  • Plan a CRAFT-structured IoT lesson for their classroom

Key Activities

  • Simulator warm-up in makecode.microbit.org (no hardware needed)
  • Role A: Sensor Node program โ€” read a sensor, radio.send the value
  • Role B: Aggregator Node program โ€” radio.receive, running average/min/max, LED display
  • Paired breakout: one participant builds Role A, partner builds Role B, share a radio group
  • Level-up: multi-sensor mesh (2+ senders โ†’ 1 aggregator) or MicroPython port
  • Sensor verification in simulator (known-good inputs, edge cases) + discussion of real-hardware calibration surprise
  • IoT lesson template customization (two-node design)

Talk:Do Ratio

~30 min facilitator-led / ~150 min participant activities (1:5.0)

Hands-on (83%) Facilitator-led (17%)

Participant Takeaways

  • BBC micro:bit V2 kit โ€” SHIPPED AFTER the session to participants who complete both surveys + submit a draft IoT lesson (yours to keep)
  • MakeCode .hex files for Sensor Node + Aggregator Node (ready to flash when hardware arrives)
  • MicroPython radio starters (sender + receiver)
  • Sensor reference guide
  • NGSS alignment crosswalk
  • Two-node IoT lesson template

Date: April 25, 2026 ยท 8:30 AM โ€“ 12:00 PM PST ยท Virtual (Zoom)

Zoom: Join Workshop 3 (Password is in your calendar invitation.)

Focus: Edge Computing + IoT as a two-device conversation โ€” sensor node โ†” aggregator node โ€” built entirely in the MakeCode online simulator.

Talk:Do Ratio: ~30 min facilitator / ~150 min participant activities (1:5.0)

Hardware: None required for the live session. Everything runs in makecode.microbit.org in your browser. Participants who complete both surveys + submit a draft IoT lesson receive a BBC micro:bit V2 kit in the mail โ€” yours to keep, to bring this into your classroom.

Take computing out of the browser โ€” conceptually โ€” while still teaching it in one. The MakeCode simulator gives every participant a fully-functional micro:bit inside a browser tab, and the moment you drop in a radio block, MakeCode spawns a second simulator so you can build and test a two-node IoT system without owning any hardware. That same code flashes to a real device the day your kit arrives.

Surveys

  • ๐Ÿ“ Pre-Survey โ€” complete at the start of the session
  • ๐Ÿ“Š Post-Survey โ€” complete at the end of the session
  • ๐Ÿ“ฆ Completing both surveys + submitting a draft IoT lesson unlocks your mailed micro:bit V2 kit.

Shared Workspace

Pre-Session Requirements

  • A modern browser (Chrome or Edge recommended) with makecode.microbit.org accessible
  • An LLM of choice open in a second tab (ChatGPT, Claude, or Gemini)
  • A pre-session email will confirm your browser can load MakeCode and that both simulators render
  • No hardware needed. If your shipped kit has already arrived, great โ€” but you do not need it for any activity today.

Learning Objectives

  • Explain Edge Computing and IoT and why the two-role pattern (sensor + aggregator) is the backbone of real-world IoT
  • Build a Sensor Node program that reads a sensor and broadcasts the value on a radio group
  • Build an Aggregator Node program that receives values, aggregates them (running average, min/max, counts, thresholds), and displays a result
  • Pair two participants on the same radio group so one’s sender talks to the other’s receiver โ€” all in simulation
  • Use an LLM to assist with writing, debugging, and extending both halves of the system
  • Optionally level-up from MakeCode (blocks) to MicroPython (text) โ€” still in a simulator
  • Align a two-node sensor activity to specific NGSS performance expectations and CTE pathways
  • Identify CRAFT phases in the session and plan how to replicate the structure

Session Resources

MicroPython Starter Scripts

Ready-to-flash equivalents of the simulator activities โ€” run in the python.microbit.org simulator today, flash to your kit when it arrives:

Programming Environments

  • MakeCode Editor โ€” block-based, runs two simulated micro:bits side-by-side when you use radio (primary)
  • MicroPython Editor โ€” text-based, also has an online simulator (level-up)

Key Activities

CRAFT PhaseActivityDurationType
โ€”Opening: Simulator Check + Icebreaker + Pre-Survey15 minYou Do
ContextualizeCRAFT Orientation + IoT as a Two-Device Conversation15 minYou Do
ReframePoll + “Physical Computing Lives in the Browser Too”15 minListen
ReframeBreakout: Barriers to Teaching Hardware Without Hardware15 minYou Do
AssembleServer Room Guardian: live-build sensor + aggregator, then swap in your own sensor30 minYou Do
AssembleEnhancements + Brainstorm: liveness ping, Data panel graphing, LLM-driven extensions30 minYou Do
AssembleLevel Up: Multi-Sensor Mesh (A) or MicroPython Port (B)30 minYou Do
FortifyExperiment: Verify Simulated Data + Calibration Preview15 minYou Do
TransferBuild: Two-Node IoT Lesson + Pair-Share + Post-Survey15 minYou Do

The Two-Role IoT Pattern

Every IoT system โ€” a smart farm, a weather station, a wearable, a traffic monitor โ€” is at its heart a conversation between two kinds of devices:

RoleJobWhat It Looks Like in MakeCode
Sensor NodeRead the world. Broadcast the reading.on start: radio.setGroup(N) ยท forever: radio.sendNumber(input.temperature()) ยท sleep ยท repeat
Aggregator NodeListen to one or many senders. Aggregate. Display. Decide.on start: radio.setGroup(N) ยท on radio.receivedNumber(v): update running average, show on LEDs, trigger alert

When you drag a radio block into MakeCode, the simulator automatically shows two simulated micro:bits so you can watch the message go from sender to receiver in real time. That’s your classroom-ready IoT demo โ€” no mailing list required.

Level-Up Tracks

Track A โ€” Multi-Sensor Mesh: Add a second sensor node by opening your Sensor project in a second browser tab (MakeCode pairs two simulators per tab, but tabs on the same origin share the radio channel). Two senders broadcast on the same group with different IDs; the aggregator tracks per-sender averages and flags which node is out of range.

Track B โ€” MicroPython Port: Port your Sensor Node and Aggregator Node to MicroPython using python.microbit.org. Ask an LLM to translate your blocks. Run both scripts in the MicroPython simulator and confirm the message still goes through.

Participant Takeaways

  • MakeCode project files (.hex) for Sensor Node + Aggregator Node + level-up extensions
  • MicroPython radio starter scripts
  • Sensor reference guide for all micro:bit V2 onboard sensors
  • NGSS alignment crosswalk (IoT activities mapped to performance expectations)
  • Customized two-node IoT lesson template with student-facing CtM prompt
  • Digital Toolkit PDF with all resources
  • BBC micro:bit V2 kit mailed after the session โ€” for participants who complete both surveys and submit a draft IoT lesson

Reframe Theme

“IoT is a CONVERSATION between devices โ€” and you can have that conversation in a browser tab.” โ€” The MakeCode simulator runs two full micro:bits side-by-side the moment you add a radio block. Your students don’t need hardware on day one. You don’t need hardware today. When the kit arrives, the same .hex flashes to the real device โ€” the code doesn’t change.

Acknowledgement

This material is based upon work supported by the National Science Foundation award #2230997. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation.


Other Workshops: Workshop 1: AI for STEM ยท Workshop 2: Verifying AI Outputs

โ† Back to Series Overview ยท โ† All Professional Development

You just built a two-node IoT system in a browser, with AI as your co-pilot, and verified it like an engineer. When your kit arrives, the same .hex files flash to real hardware. Your students can do this too โ€” even in schools with zero hardware budget.