updated framing

Author: Mister-Raggs

_config.yml

@@ -141,7 +141,7 @@ relative_links:
 
 titles_from_headings:
   strip_title: true
-  collections: true
+  collections: false
 
 compress_html:
   comments: ["<!--", "-->"]

about.md

@@ -16,13 +16,19 @@ accent_image: /assets/images/mountain.jpg
 
 I'm a software engineer drawn to operational problems — the kind where something is slow, manual, or breaking under load. Most of my work has been finding those points and building systems that fix them structurally: faster log search, automated on-call workflows, distributed ingestion pipelines. Lately that's been pulling me toward AI systems specifically — where the same reliability and latency problems show up, but the infrastructure is less mature.
 
+## What I Work On
+
+I build backend systems, data pipelines, and infrastructure tooling — and increasingly, I build them with AI at the core. My recent work sits at the intersection: LLM-powered log anomaly detection, autonomous agents that manage cloud infrastructure, voice-driven code generation pipelines, and RAG systems for real-world retrieval problems.
+
+I care about the parts most people skip — how you evaluate LLM output in production, how you make an agent reliable enough to run unsupervised, how you design pipelines that fail gracefully when the model doesn't.
+
 ## Experience
 
 ### Amazon
 Built incident-resolution tooling and a fast log search microservice; integrated log retrieval into developer chat workflows.
 
 ### UC San Diego
-Designed hands-on assignments covering RAG pipelines, multi-agent orchestration, and LLM fine-tuning.
+Designed hands-on coursework covering RAG pipelines, multi-agent orchestration, and LLM fine-tuning for 200+ students.
 
 ### AARK Global
 Orchestrated high-throughput document pipelines and built conversational AI over indexed data.
@@ -31,18 +37,18 @@ Orchestrated high-throughput document pipelines and built conversational AI over
 Automated sentiment analysis and scaled ingestion pipelines for realtime analytics.
 
 ### Research
-Developed BioBERT models and LLM fine-tuning work for clinical decision support.
+Fine-tuned BioBERT models for clinical decision support, achieving 91.6% routing accuracy.
 
 ## Skills
 
-- **Distributed Systems:** Large-scale data pipelines, log search, low-latency microservices
+- **Distributed Systems & Infrastructure:** Data pipelines, log search, low-latency microservices, AWS
 - **Observability:** Incident tooling, operational visibility, reliability engineering
-- **ML/AI for Infrastructure:** Anomaly detection, RAG, LLM fine-tuning, operational intelligence
-- **Backend:** Python, FastAPI, microservices, AWS
+- **AI/ML Engineering:** LLM pipelines, RAG, fine-tuning, agentic systems, LLM evaluation
+- **Backend:** Python, FastAPI, microservices
 
 ## Interests
 
+- Applying LLMs and agents to infrastructure and operational problems
 - Distributed data pipelines and stream processing
 - Observability and operational visibility at scale
-- ML/AI applied to infrastructure reliability
-- Platforms that simplify large-scale data and AI systems
+- Making AI systems reliable enough for production