Things we’ve figured out
Notes from our engineers on problems we had to solve, from moving between clouds to testing AI output. Written down so the next person does not have to work it out again.
Migrate data from a GCP bucket to AWS
The steps for moving a storage bucket from Google Cloud to Amazon S3: assessing the data, preparing the target, transferring, and verifying the result.
On cloud migration.
QA in evaluating LLM output
Language models give a different answer each time, so pass-or-fail testing falls short. How we check model output for accuracy, consistency, format and bias.
On AI quality.
GCP tools to secure CI/CD pipelines
Google Cloud’s built-in services for securing a delivery pipeline, from vulnerability scanning of artifacts to controlling what is allowed to deploy.
On delivery pipelines.
Cloud migration tools
How to choose between a cloud provider’s own migration tools, third-party products and open-source options when moving workloads between platforms.
On cloud migration.
AWS DataSync — migrating data from GCP to AWS
A walkthrough of AWS DataSync for moving data from Google Cloud Storage to AWS, from the access key and agent setup to running the transfer task.
On cloud migration.
Stuck on something like this?
If one of these notes describes your week, tell us where you are with it.