<?xml version="1.0" encoding="utf-8"?><feed xmlns="http://www.w3.org/2005/Atom" xml:lang="en"><generator uri="https://jekyllrb.com/" version="4.3.3">Jekyll</generator><link href="https://taliaora.github.io/feed.xml" rel="self" type="application/atom+xml"/><link href="https://taliaora.github.io/" rel="alternate" type="text/html" hreflang="en"/><updated>2026-03-09T11:07:39+00:00</updated><id>https://taliaora.github.io/feed.xml</id><title type="html">Natalia M. Sikora</title><subtitle>A PhD Candidate, EPSRC Postdoctoral Fellow and machine learning researcher specialising in multimodal data fusion, explainable AI (XAI), and biomedical AI. Swansea, Oxford, Nottingham.</subtitle><entry><title type="html">Crushing ML Latency: The (Un)Official Best Practices for Systems Optimisation</title><link href="https://taliaora.github.io/blog/2025/crushing-ml-latency-the-unofficial-best-practices-for-systems-optimisation/" rel="alternate" type="text/html" title="Crushing ML Latency: The (Un)Official Best Practices for Systems Optimisation"/><published>2025-11-04T10:43:23+00:00</published><updated>2025-11-04T10:43:23+00:00</updated><id>https://taliaora.github.io/blog/2025/crushing-ml-latency-the-unofficial-best-practices-for-systems-optimisation</id><content type="html" xml:base="https://taliaora.github.io/blog/2025/crushing-ml-latency-the-unofficial-best-practices-for-systems-optimisation/"><![CDATA[]]></content><author><name></name></author></entry></feed>