AGT
Artificial Intelligence Agent
Agentic workflow, planning, tool use, memory, reflection, multi-agent collaboration, and deep research automation.
ReActPlanningMemoryMulti-Agent
AI Research Engineer
AI Research Laboratory
Exploring intelligence from deep learning to autonomous agents through open source systems, paper reading, and reusable AI engineering knowledge.
LLMAgentRAGMultimodal AIComputer Vision
AI
LLM
Agent
RAG
CV
RL
Research Graph Seed
Projects, papers, concepts, and implementation notes will become clickable graph nodes in Phase 2.
Research Direction
LLM agents, RAG systems, multimodal learning, computer vision, and AI engineering foundations.
Open Source Systems
Agent simulators, deep search workflows, fine-tuning pipelines, CV projects, and ML implementations.
Educational Platform
A living map of concepts, papers, code paths, and learning outcomes for AI research growth.
Research Focus
A personal map from models to autonomous systems.
The site organizes learning, research projects, implementation notes, and paper reading around the concepts that matter for practical AI research.
LLM
Large Language Model Engineering
Fine-tuning pipelines, SFT, LoRA, QLoRA, DPO, evaluation, prompt systems, and model deployment practice.
SFTLoRAQLoRADPO
RAG
Retrieval-Augmented Generation
Knowledge retrieval, query rewriting, indexing, ranking, GraphRAG direction, and production RAG evaluation.
RetrieverRerankGraphRAGEvaluation
MM
Multimodal AI
Vision-language reasoning, OCR, image understanding, video restoration, and multimodal representation learning.
VLMOCRVideoVision-Language
CV
Computer Vision
Image restoration, segmentation, detection, OCR pipelines, visual representation, and applied CV research systems.
RestorationOCRDetectionRepresentation
ML
Machine Learning Foundation
Bayesian networks, optimization algorithms, reinforcement learning, and framework-level implementation practice.
BayesianOptimizationRLFramework
Research Timeline
A staged path from foundations to autonomous AI systems.
The roadmap is intentionally research-oriented: build foundations, implement systems, then connect papers and projects into a reusable knowledge graph.
2024
1
Deep Learning Foundation
Build mathematical foundations, machine learning algorithms, deep learning implementation habits, and optimization intuition.
PythonMathMLDeep Learning
2025
2
Computer Vision and Multimodal
Move from model understanding to applied perception systems: OCR, video restoration, multimodal learning, and visual reasoning.
CVOCRVideoMultimodal
2026
3
LLM Agent and Autonomous AI System
Focus on LLM agents, RAG, tool use, deep search, multi-agent systems, and agentic research workflows.
LLMAgentRAGAI System
Project System
Research projects as reusable knowledge artifacts.
Each project is stored as Markdown metadata today and can become a detailed architecture, paper, and code explanation page in Phase 2.
Agent Systems
Agent-You-MustKnows
A structured knowledge base for core AI agent concepts, research papers, and engineering patterns.
Agent Systems
CognitiveTemp DeepSearch Agents
A deep search agent direction for decomposing questions, retrieving evidence, reasoning over sources, and generating research-grade answers.
Agent Systems
NEXUS
An AI agent system exploration project focused on emergent workflows, coordination, and agent environment design.
Agent Systems
NEXUS Navigating Emergent X-agent Universe Simulator
A simulator direction for studying emergent multi-agent behavior, world state, and autonomous interaction loops.