Modern Software Development Life Cycle: From Waterfall to Agentic CD
An architectural audit of modern software development life cycles, evaluating trunk-based delivery, shift-left testing, DORA metrics, and autonomous agent loops.
Fodev JEO
LATEST INVESTIGATIONS
Source audits & field notes
Google Technical Interviews: Systems Architecture, Algorithmic Rigor, and Googleyness
A source audit evaluating Google technical interviews, analyzing algorithmic problem-solving rubrics, planet-scale system design, and AI-era hiring transformations.
Unifying Context Reranking and Generation: Analyzing NVIDIA's RankRAG Architecture
A source audit analyzing NVIDIA’s RankRAG, evaluating dual-purpose instruction tuning for context ranking and answer generation across dense retrieval pipelines.
Agentic RAG Architectures: Query Reformulation, Self-Correction, and CRAG
A source audit analyzing Agentic Retrieval-Augmented Generation, comparing vanilla RAG limitations, Corrective RAG (CRAG), Self-RAG, and Graph-RAG systems.
The Architecture of XaaS: Shared Responsibility, Serverless, and GPUaaS
A technical source audit analyzing XaaS service boundaries, comparing IaaS, PaaS, and FaaS with modern Model-as-a-Service and GPU-as-a-Service infrastructure.
FunAudioLLM Architecture: SenseVoice, CosyVoice, and Duplex Speech Systems
A source audit analyzing Alibaba’s FunAudioLLM framework: SenseVoice non-autoregressive ASR, CosyVoice flow-matching speech synthesis, and duplex voice agents.
Benchmarking Vector Databases for LLMs: Architecture, Indexing, and Trade-Offs
An architectural source audit of vector databases for LLM systems, evaluating HNSW vs. IVF indexing, memory footprints, filtering overhead, and storage engines.
GraphRAG Architecture: Hierarchical Knowledge Graphs and Global Corpus Summarization
A source audit evaluating Microsoft GraphRAG, analyzing entity-relation extraction, Leiden community detection, and global dataset summarization trade-offs.
Intelligent Query Routing with RouteLLM: Cost-Quality Pareto Frontiers and Routing Models
A source audit analyzing UC Berkeley’s RouteLLM framework, evaluating matrix factorization, BERT classifiers, LLM judges, and cost-quality Pareto frontiers.
High-Performance Python Data Analytics: Arrow, Polars, and DuckDB Mechanics
A source audit analyzing high-performance Python data analytics, comparing Apache Arrow memory layout, Polars multi-threading, and DuckDB vectorized execution.
인기 태그
Search article titles, categories, and tags. Full text is searched when needed.
Type to search articles.