LiftLoop — training coaching on Alexa+

Inspiration

What it does

How we built it

Challenges we ran into

Accomplishments that we're proud of

What we learned

What's next for LiftLoop — training coaching on Alexa+

Inspiration

Every home lifters we spoke to described the same thing: generic programs ignore how the last session actually went. Building for Alexa+ meant the assistant could sit inside voice rather than asking people to open yet another app.

What it does

LiftLoop takes coaching strength sessions week by week end to end. The planning step builds a concrete plan; a memory step remembers what worked and what did not, so repeat suggestions stop; a scheduled job re-runs the plan on a cadence and pushes the output to where it is needed. Everything is reachable through voice-first conversations.

How we built it

The agent is written in Python with Strands Agents SDK and runs on AWS Lambda, calling Amazon Bedrock for the reasoning steps. State lives in DynamoDB, EventBridge and API Gateway glue the scheduled and event paths together, and the whole stack deploys from a single CDK/CloudFormation app.

Challenges we ran into

Keeping the plan coherent when memory changed the constraints mid-decision was the hard part — an early build happily produced plans that contradicted themselves. Cold starts were the other issue: anything on the voice-first conversations path has to feel instant, which pushed us to pre-warm the Amazon Bedrock client and shrink the tool graph.

Accomplishments that we're proud of

It works unattended. We left it running for a week and it kept producing useful output without anyone touching it — that was the moment it stopped being a demo. It also costs almost nothing to run, which matters for home lifters.

What we learned

Strands Agents SDK makes it easy to attach real state to an agent; the hard engineering is conflict resolution between tools once they disagree. We also learned how much of perceived quality in a Alexa+ experience comes down to latency rather than model choice.

What's next for LiftLoop

Next we want shared use so two people can edit the same plan, a cost estimate driven by live pricing, and a second surface — moving from voice to a companion view so the same agent is usable in more than one room.

Built With

• amazon-alexa
• amazon-bedrock
• aws-lambda
• fitness
• liftloop
• strands-agents-sdk