455 CARDS · 41 WEEKS · EVERY SOURCE THE PLAN NAMES

Everything you could
study, filtered.

Every video, chapter, paper and repository in the programme, as one card each, with the time it costs and why it is there. Filter by how long you have.

Type Watch Read Run Track Biology AI & stats Build Craft Depth Core Optional Time ≤15 ≤30 ≤60 Week This week Clear all

8/12 Run 30 min W34 Build

vibrantlabsai/ragas

Compute context recall and faithfulness on 10 of your Q&A; compare with your human labels before trusting it

Open ↗

In: LABLOOP-W34 Notebook

9/12 Run 30 min W34 Build

run-llama/llama_index

Only as a reference implementation of chunking and citation; keep your own pipeline small

Open ↗

In: LABLOOP-W34 Notebook

10/12 Run 30 min W34 Build

illuin-tech/colpali

Try page-image retrieval on 20 scanned SOP pages; compare recall@5 with text chunks

Open ↗

In: LABLOOP-W34 Notebook

11/12 Run 30 min W34 Build

UKGovernmentBEIS/inspect_ai

Wrap the 60 Q&A as an eval with a model-graded scorer and your human labels

Open ↗

In: LABLOOP-W34 Notebook

12/12 Run 30 min W34 Build

ggml-org/llama.cpp

Serve the embedding model locally on the mini PC for the on-prem story

Open ↗

In: LABLOOP-W34 Notebook