AI tutor, macaque templates, reliability calibration + free open-source tutor default - #253

Open

ataha24 wants to merge 21 commits into

masterfrom

enh/ai-tutor-learning-mode

Commits on Jul 23, 2026

ENH: visual overhaul, macaca templates, and new validation features

ataha24
authored and
Dhananjhay
committed
8cfe1fc
ENH: add AI tutor guided learning mode (/learn)

ataha24
authored and
Dhananjhay
committed
152d03b
add new feature for various LLMs

ataha24
authored and
Dhananjhay
committed
c3af7de
add support for full afid set and improve the visualization settings

ataha24
authored and
Dhananjhay
committed
061a74e
ENH: bring-your-own-key tutor, RAG grounding, and rater-reliability calibration

ataha24
authored and
Dhananjhay
committed
07029c4
Add inter-template + rater-reliability analysis, figures, and Aperture education manuscript

ataha24
authored and
Dhananjhay
committed
182a72e
Elevate paper figures: live-app hero, glass-brain landmarks, graphical abstract

ataha24
authored and
Dhananjhay
committed
9f8aa09
style: apply black/isort/ruff to satisfy the CI lint gate

ataha24
authored and
Dhananjhay
committed
cfa5625
Reframe reference-data analysis: difficulty benchmark + worked QC catch

ataha24
authored and
Dhananjhay
committed
510ad05
Polish manuscript for submission: proofread, ethics + CRediT, naming

ataha24
authored and
Dhananjhay
committed
940847b
Finalize author block: rebuild affiliations and CRediT roles

ataha24
authored and
Dhananjhay
committed
bcb211b
Tighten figures for publication: field-guide Fig 3, drop titles, simpler cites

ataha24
authored and
Dhananjhay
committed
61feb1d
Remove scratch Fig 3 iteration renders from paper_figures/

ataha24
authored and
Dhananjhay
committed
c5cb668
Enable free open-source tutor default (Groq Llama 3.3 70B)

ataha24
authored and
Dhananjhay
committed
3c71841
Fix migration fork: parent RAG migration on the real head

ataha24
authored and
Dhananjhay
committed
f4ddc20
Regenerate poetry.lock to include openai dependency

ataha24
authored and
Dhananjhay
committed
8296842
Modernize CI/deploy Python: drop EOL 3.8, pin deploy to 3.11

ataha24
authored and
Dhananjhay
committed
5a59a6c
Default guided-learning tutor to Groq Llama 3.3 70B

ataha24
authored and
Dhananjhay
committed
84137b3
Fix rater percentile: report precision (higher = better)

ataha24
authored and
Dhananjhay
committed
7cc3aed
Add Back button to guided-learning mode

ataha24
authored and
Dhananjhay
committed
d8d10d0
Add median + IQR to validator session summary

ataha24
authored and
Dhananjhay
committed
dd1b18f

Navigation Menu

Uh oh!

AI tutor, macaque templates, reliability calibration + free open-source tutor default - #253

AI tutor, macaque templates, reliability calibration + free open-source tutor default#253
ataha24 wants to merge 21 commits into
masterfrom
enh/ai-tutor-learning-mode

Commits on Jul 23, 2026

ENH: visual overhaul, macaca templates, and new validation features

ENH: add AI tutor guided learning mode (/learn)

add new feature for various LLMs

add support for full afid set and improve the visualization settings

ENH: bring-your-own-key tutor, RAG grounding, and rater-reliability calibration

Add inter-template + rater-reliability analysis, figures, and Aperture education manuscript

Elevate paper figures: live-app hero, glass-brain landmarks, graphical abstract

style: apply black/isort/ruff to satisfy the CI lint gate

Reframe reference-data analysis: difficulty benchmark + worked QC catch

Polish manuscript for submission: proofread, ethics + CRediT, naming

Finalize author block: rebuild affiliations and CRediT roles

Tighten figures for publication: field-guide Fig 3, drop titles, simpler cites

Remove scratch Fig 3 iteration renders from paper_figures/

Enable free open-source tutor default (Groq Llama 3.3 70B)

Fix migration fork: parent RAG migration on the real head

Regenerate poetry.lock to include openai dependency

Modernize CI/deploy Python: drop EOL 3.8, pin deploy to 3.11

Default guided-learning tutor to Groq Llama 3.3 70B

Fix rater percentile: report precision (higher = better)

Add Back button to guided-learning mode

Add median + IQR to validator session summary