Linux Operating System Architecture: Kernel Rings, System Calls, and Virtual Memory
A source audit evaluating Linux operating system internals, analyzing kernel space separation, system call dispatching, virtual memory paging, and boot sequencing.
Fodev JEO
LATEST INVESTIGATIONS
Source audits & field notes
Kubernetes for Machine Learning: GPU Scheduling, NVIDIA NIM, and KEDA Autoscaling
A source audit analyzing Kubernetes in MLOps, evaluating GPU topology scheduling, NVIDIA NIM microservice deployment, and KEDA event-driven autoscaling.
Apple Intelligence Architecture: On-Device AFM and MDLOO Alignment
An architectural audit of Apple Intelligence Foundation Models, analyzing 3B on-device Palettization, dynamic LoRA adapters, MDLOO alignment, and Private Cloud Compute.
The Production LLM Stack: Architectural Blueprint Across Seven Engineering Layers
A source audit analyzing the open-source production LLM landscape, evaluating data ingestion, training frameworks, serving engines, evals, and guardrails.
Optimizing Production RAG: Automated Hyperparameter Tuning and Evaluation Harnesses
A source audit evaluating automated RAG optimization, analyzing combinatorial hyperparameter tuning across chunking, embeddings, rerankers, and Ragas metrics.
Reinforcement Learning from Human Feedback: Mechanics of PPO, DPO, and Reward Modeling
A source audit evaluating RLHF in large language models, analyzing reward modeling, PPO policy gradients, Direct Preference Optimization, and reward hacking limits.
Graph-Structured Speculative Decoding: Accelerating LLM Inference via DAG Verification
A source audit analyzing Graph-Structured Speculative Decoding, evaluating DAG-based multi-candidate verification, tree attention, and latency reduction in LLMs.
Auditing Long-Context LLMs: NeedleBench, Bilingual Reasoning, and Depth Degradation
A source audit analyzing Shanghai AI Lab’s NeedleBench, evaluating multi-needle retrieval, bilingual reasoning, Ancestral Trace Challenges, and depth degradation.
Data Quality in Machine Learning Systems: Contracts, Drift, and Feat Stores
An architectural audit of machine learning data quality, analyzing data contracts, streaming Flink validation, train-serve skew, and distribution drift metrics.
Orchestrating AI Agents: From LangChain LCEL Foundations to LangGraph State Machines
A technical source audit of LangChain agent orchestration, comparing LCEL declarative chains, ReAct decision loops, and LangGraph cyclical state machines.
인기 태그
Search article titles, categories, and tags. Full text is searched when needed.
Type to search articles.