Add 2026 fellow profile for Ava Faubus (#2881)

* Create avafaubus.md

* Update avafaubus.md

* Add files via upload

* Rename ava-faubus.png to Ava-Faubus.png

* Delete assets/images/team/fellows-2026/Ava-Faubus.png

* Add files via upload

* style: pre-commit fixes

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

Author: avafaubus

_fellows/2026/avafaubus.md

@@ -0,0 +1,28 @@
+---
+layout: fellow
+pagetype: fellow
+shortname: avafaubus
+permalink: /fellows/avafaubus.html
+fellow-name: Ava Faubus
+title: Ava Faubus - IRIS-HEP Fellow
+active: True
+dates:
+  start: 2026-06-08
+  end: 2026-08-15
+photo: /assets/images/team/fellows-2026/ava-faubus.jpg
+institution: Wellesley College
+e-mail: af214@wellesley.edu
+focus-area: ia
+challenge-area:
+project_title: Unsupervised Machine Learning for Muon data quality monitoring in CMS
+project_goal:
+    This project develops and evaluates unsupervised machine learning methods to improve automated data quality monitoring (AutoDQM) for the Cathode Strip Chambers (CSCs) in CMS. As CMS enters the high-luminosity era, scalable and reliable anomaly detection is essential for maintaining detector performance while reducing the need for manual inspection. Principal component analysis (PCA) and autoencoder models are investigated as alternatives to the current reference-run-based statistical approach, with the goal of reducing false anomaly detections by AutoDQM. Models will be trained with lumisection granularity, and their performance will be evaluated against the current statistical methods present in AutoDQM.
+mentors:
+  - Indara Suarez (Boston University)
+  - Frank Golf (Boston University)
+proposal: assets/pdf/fellows-2026/USA002-proposal-Ava-Faubus.pdf
+presentations: []
+current_status: ""
+github-username: avafaubus
+linkedin-profile:
+---