jidonglab

1 project a week. Building and sharing the entire process — from idea to shipped product in 7 days. Currently: AI news automation.

korea Joined on Feb 25, 2026 jee599@naver.com https://jidonglab.com/

Pinned

jidonglab

Mar 19

Claude Code Agent Teams Can Spawn Agents. It Just Doesn't Know Which Ones to Use.

#ai #claude #productivity #opensource

4 min read

jidonglab

Mar 6

What 100 Trending GitHub Projects Tell Us About Where AI Is Actually Going

#ai #opensource #productivity #webdev

4 min read

jidonglab

Mar 3

Anthropic Dropped 13 Free Courses — I Broke Down Every Single One

#ai #productivity #claude #beginners

3 min read

jidonglab

Aug 4

Why Repetition Penalty Breaks JSON and Code Generation

7 min read

Want to connect with jidonglab?

Create an account to connect with jidonglab. You can also sign in below to proceed if you already have an account.

Create Account

Already have an account? Sign in

jidonglab

Aug 3

RoPE Scaling: Why Raising rope_theta Breaks Short Context

7 min read

jidonglab

Aug 3

Claude Prompt Caching: Why cache_read_input_tokens Stays 0

7 min read

jidonglab

Aug 2

Attention Sinks: Why Evicting Token 0 Wrecks Sliding-Window KV

7 min read

jidonglab

Aug 2

Why temperature=0 Still Gives Different Answers: Batch Invariance

7 min read

jidonglab

Aug 1

Speculative Decoding: Why 80% Acceptance Still Loses at Batch 64

8 min read

jidonglab

Aug 1

Why JSON Schema Field Order Breaks Structured Output Accuracy

#ai #llm #machinelearning #python

7 min read

jidonglab

Jul 31

Why INT4 Weight-Only Quantization Doesn't Speed Up Prefill

#llm #machinelearning #performance

7 min read

jidonglab

Jul 31

DPO Likelihood Displacement: Why Chosen Responses Get Rarer

7 min read

jidonglab

Jul 30

ColBERT Late Interaction: Why MaxSim Beats One Vector Per Chunk

7 min read

jidonglab

Jul 30

Surface Form Competition: Why Log-Prob Answer Scoring Fails

7 min read

jidonglab

Jul 29

Activation Outliers: Why W8A8 INT8 Quantization Needs SmoothQuant

8 min read

jidonglab

Jul 29

Cross-Encoder Reranker Score Calibration: Why 0.5 Cutoffs Fail

8 min read

jidonglab

Jul 28

MCP Tool Sprawl: Why 40 Tools Wreck Tool Selection Accuracy

7 min read

jidonglab

Jul 28

Why I measure interview silence with WebAudio, not the Speech API

#webaudio #javascript #webdev #ai

5 min read

jidonglab

Jul 28

Best-of-N Sampling: Why N=64 Scores Higher and Answers Worse

7 min read

jidonglab

Jul 28

Building an AI mock interviewer that tells you why you'd fail: what the turn engine taught me

#ai #webdev #javascript #career

5 min read

jidonglab

Jul 27

Reciprocal Rank Fusion: Why k=60 Buries Your Best Hit

7 min read

jidonglab

Jul 27

Hard Negative Mining Breaks Embedding Fine-Tunes Without Denoising

7 min read

jidonglab

Jul 26

LLM Eval Noise: Why Your 2-Point Accuracy Win Isn't Real

7 min read

jidonglab

Jul 26

Grouped Query Attention: Why 8 KV Heads Decide Your Batch Size

7 min read

jidonglab

Jul 25

Binary Quantized Embeddings: 32x Smaller, If You Center First

8 min read

jidonglab

Jul 25

MoE Capacity Factor: Why Mixture-of-Experts Drops Your Tokens

#deeplearning #llm #machinelearning #performance

8 min read

jidonglab

Jul 24

Hubness in Vector Search: Why One Chunk Tops Every RAG Query

7 min read

jidonglab

Jul 24

Min-p Sampling: Why top_p Truncates the Wrong Tail

6 min read

jidonglab

Jul 23

Chunked Prefill: Why One Long Prompt Stalls Every Decode

8 min read

jidonglab

Jul 22

Sequence Packing Leaks Across Documents Unless You Mask It

6 min read

jidonglab

Jul 21

KV Cache Quantization: Why Keys and Values Need Different Axes

6 min read

jidonglab

Jul 21

Multi-LoRA Serving: Why 100 Adapters Fit on One GPU

7 min read

jidonglab

Jul 20

Token Healing: Why a Trailing Space Wrecks LLM Completions

7 min read

jidonglab

Jul 20

Digit Tokenization: Why Commas Fix LLM Arithmetic

#ai #algorithms #llm #nlp

8 min read

jidonglab

Jul 19

Matryoshka Embeddings: Truncate Vector Dimensions, Keep Recall

7 min read

jidonglab

Jul 19

Why Filtered Vector Search Quietly Destroys HNSW Recall

7 min read

jidonglab

Jul 18

LLM-as-Judge Position Bias: Measure It Before You Ship

8 min read

jidonglab

Jul 18

YaRN vs NTK-Aware RoPE Scaling: Why Long Context Breaks

7 min read

jidonglab

Jul 17

Prompt Caching: How One Dynamic Token Kills the 90% Discount

7 min read

jidonglab

Jul 17

Attention Sinks: Why Streaming LLMs Break When You Evict Token 0

#deeplearning #llm #machinelearning #nlp

6 min read

jidonglab

Jul 16

Why Temperature 0 Doesn't Make Your LLM Deterministic

6 min read

jidonglab

Jul 16

Speculative Decoding: Why a Great Draft Model Still Caps Speedup

6 min read

jidonglab

Jul 15

Constrained Decoding: Force Valid JSON Without Wrecking Accuracy

6 min read

jidonglab

Jul 13

GRPO Explained: Why DeepSeek Dropped the Critic in RLHF

6 min read

jidonglab

Jul 12

Online Softmax: How FlashAttention Skips the N N Matrix

7 min read

jidonglab

Jul 12

Late Interaction Retrieval: Why ColBERT Beats Single-Vector RAG

7 min read

jidonglab

Jul 11

Why repetition_penalty Quietly Corrupts Your Code Generation

6 min read

jidonglab

Jul 11

DPO Likelihood Displacement: When Preferred Answers Get Rarer

6 min read

jidonglab

Jul 10

Why Token Logprobs Beat Asking Your LLM How Confident It Is

7 min read

jidonglab

Jul 9

Multi-Token Prediction: DeepSeek's Built-In Draft Model

6 min read

jidonglab

Jul 9

AWQ: How Activation-Aware Quantization Saves 4-bit LLMs

6 min read

jidonglab

Jul 8

PagedAttention: Why Static Batching Wastes Your KV Cache

6 min read

jidonglab

Jul 8

11 Claude Agents Audited My Kit. 2 'Major Bugs' Were Fake

#ai #claudecode #opensource #programming

6 min read

jidonglab

Jul 8

Semantic Entropy: Detect LLM Hallucinations Without Ground Truth

6 min read

jidonglab

Jul 7

Contextual Retrieval: Fix the RAG Chunk That Lost Its Context

6 min read

jidonglab

Jul 7

Prefilling Claude's Response: Steer Output Without JSON Mode

6 min read

jidonglab

Jul 7

I Cut My Claude Code Prompt Overhead 84% Per Turn (v1.4)

7 min read

jidonglab

Jul 6

Multi-head Latent Attention: The KV Cache Trick Beyond GQA

7 min read

jidonglab

Jul 6

Why Your Mixture-of-Experts Model Silently Drops Tokens

7 min read

jidonglab

Jul 5

Embedding Anisotropy: Why Cosine Similarity Never Hits Zero

6 min read

jidonglab

Jul 5

Return Claude's Thinking Blocks or Your Agent Breaks

6 min read

jidonglab

Jul 4

Binary Quantized Embeddings: 32x Smaller Vectors, Recall Intact

7 min read

jidonglab

Jul 4

Prefill/Decode Disaggregation: Stop Serving LLMs on One GPU

6 min read

2 Week Community Wellness Streak

1 Week Community Wellness Streak

Writing Debut

Want to connect with jidonglab?