DEV Community

jidonglab profile picture

jidonglab

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

LLM-as-Judge Position Bias: Swapping A/B Flipped 87 of 300 Verdicts

LLM-as-Judge Position Bias: Swapping A/B Flipped 87 of 300 Verdicts

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.

Already have an account? Sign in
git push --force-with-lease Is Unsafe If Your Editor Auto-Fetches

git push --force-with-lease Is Unsafe If Your Editor Auto-Fetches

1
6 min read
Python ThreadPoolExecutor Swallowed 212 Claude API Errors Silently

Python ThreadPoolExecutor Swallowed 212 Claude API Errors Silently

6 min read
AWS ALB Idle Timeout Killed 14% of My Voice WebSockets at 60s

AWS ALB Idle Timeout Killed 14% of My Voice WebSockets at 60s

2
6 min read
Measuring What My Electron App Sends: 94 Sockets, 0 Out

Measuring What My Electron App Sends: 94 Sockets, 0 Out

1
1
5 min read
vLLM Speculative Decoding: 1.8x Faster Alone, Slower at 32 Users

vLLM Speculative Decoding: 1.8x Faster Alone, Slower at 32 Users

2
1
7 min read
Claude Code Deny Rules Block Read(.env) But Not cat .env

Claude Code Deny Rules Block Read(.env) But Not cat .env

2
4
6 min read
pgvector HNSW Filtering Returned 0 Rows for 52 of My 400 Queries

pgvector HNSW Filtering Returned 0 Rows for 52 of My 400 Queries

6 min read
frequency_penalty Broke My JSON: 47 of 88 Long Invoices Lost Items

frequency_penalty Broke My JSON: 47 of 88 Long Invoices Lost Items

7 min read
Wrapping Claude Code and Codex in Electron: A Stop Bug

Wrapping Claude Code and Codex in Electron: A Stop Bug

6 min read
Docker Bypasses UFW: 3 Ports I Denied Were Open to the Internet

Docker Bypasses UFW: 3 Ports I Denied Were Open to the Internet

6 min read
Silero VAD Endpointing Cut Off 23% of Answers: 700ms Isn't a Turn

Silero VAD Endpointing Cut Off 23% of Answers: 700ms Isn't a Turn

6 min read
nginx proxy_buffering Broke My Claude SSE Stream: 0.7s to 3.4s TTFT

nginx proxy_buffering Broke My Claude SSE Stream: 0.7s to 3.4s TTFT

2
1
7 min read
BPE Tokenization: A Trailing Space in My Prompt Broke 41 of 300 Labels

BPE Tokenization: A Trailing Space in My Prompt Broke 41 of 300 Labels

6 min read
13 Research Agents Told Me My Feature Was Already Free

13 Research Agents Told Me My Feature Was Already Free

5 min read
GitHub Actions paths Filter + Required Checks: 23 PRs Stuck Pending

GitHub Actions paths Filter + Required Checks: 23 PRs Stuck Pending

6 min read
Celery visibility_timeout Ran My Claude Jobs 3.3x: 1,800 Became 5,913

Celery visibility_timeout Ran My Claude Jobs 3.3x: 1,800 Became 5,913

7 min read
JSON Schema Field Order: My vLLM Classifier Answered Before Thinking

JSON Schema Field Order: My vLLM Classifier Answered Before Thinking

6 min read
Claude Prompt Caching Hit 0 of 1,412 Calls: A Timestamp Killed It

Claude Prompt Caching Hit 0 of 1,412 Calls: A Timestamp Killed It

7 min read
Codex Image Gen Saved My White Pixels as Transparent

Codex Image Gen Saved My White Pixels as Transparent

6 min read
GitHub Actions Cache Never Updates: 9 Weeks of a Frozen mypy Cache

GitHub Actions Cache Never Updates: 9 Weeks of a Frozen mypy Cache

6 min read
Claude Batch API Returned Results Out of Order: 59% of My Labels Wrong

Claude Batch API Returned Results Out of Order: 59% of My Labels Wrong

6 min read
vLLM KV Cache Math: Why My 24GB GPU Served Only 4 Requests at Once

vLLM KV Cache Math: Why My 24GB GPU Served Only 4 Requests at Once

7 min read
Docker RUN rm Didn't Delete My .env. 3 of 14 Images Still Had It

Docker RUN rm Didn't Delete My .env. 3 of 14 Images Still Had It

6 min read
I Shipped a Friends Feature, Then Dropped It the Same Day

I Shipped a Friends Feature, Then Dropped It the Same Day

1
5 min read
SQLite 'database is locked' Ignored My 30s busy_timeout: 311 Failures

SQLite 'database is locked' Ignored My 30s busy_timeout: 311 Failures

1
6 min read
Claude stop_reason max_tokens: 83 Reports Silently Cut Off Mid-JSON

Claude stop_reason max_tokens: 83 Reports Silently Cut Off Mid-JSON

6 min read
Five Parallel AI Builders, One Contract, 25 Review Fixes

Five Parallel AI Builders, One Contract, 25 Review Fixes

2
6 min read
Refmade: a free desktop app for Claude Code and Codex, without the terminal

Refmade: a free desktop app for Claude Code and Codex, without the terminal

1
3 min read
BM25 vs Embeddings: Why My Vector Search Couldn't Find Error E4012

BM25 vs Embeddings: Why My Vector Search Couldn't Find Error E4012

1
1
7 min read
Why docker stop Takes 10 Seconds: PID 1 Is Ignoring Your SIGTERM

Why docker stop Takes 10 Seconds: PID 1 Is Ignoring Your SIGTERM

6 min read
Claude Code Hooks: Exit Code 1 Doesn't Block. 9 of 9 Commands Ran

Claude Code Hooks: Exit Code 1 Doesn't Block. 9 of 9 Commands Ran

2
6 min read
Reading Claude Code and Codex Session Logs: Four Traps

Reading Claude Code and Codex Session Logs: Four Traps

1
6 min read
vLLM Temperature 0 Isn't Deterministic: 23 of 200 Answers Changed

vLLM Temperature 0 Isn't Deterministic: 23 of 200 Answers Changed

1
7 min read
Whisper Hallucinations: 'Thanks for Watching' in 9% of My Transcripts

Whisper Hallucinations: 'Thanks for Watching' in 9% of My Transcripts

6 min read
Stop Trusting Agent Self-Reports: Read the Logs Instead

Stop Trusting Agent Self-Reports: Read the Logs Instead

6 min read
A Local-First Plan, Then 4 Days of Scope Creep

A Local-First Plan, Then 4 Days of Scope Creep

1
6 min read
GitHub Actions fetch-depth: 1 Made My CI Skip Tests for 41 PRs

GitHub Actions fetch-depth: 1 Made My CI Skip Tests for 41 PRs

1
7 min read
I Built an Agent Team, Then Couldn't See What It Did

I Built an Agent Team, Then Couldn't See What It Did

5 min read
Sync Anthropic Client Blocked My asyncio Loop: Voice p95 Hit 11.4s

Sync Anthropic Client Blocked My asyncio Loop: Voice p95 Hit 11.4s

7 min read
Python json.dumps ensure_ascii: 24% More Claude Tokens, Wrong Kanji

Python json.dumps ensure_ascii: 24% More Claude Tokens, Wrong Kanji

1
6 min read
Cosine Similarity Threshold 0.8: Embedding Anisotropy Broke My RAG

Cosine Similarity Threshold 0.8: Embedding Anisotropy Broke My RAG

1
7 min read
Docker Compose depends_on Won't Wait for Postgres: 11/50 Runs Failed

Docker Compose depends_on Won't Wait for Postgres: 11/50 Runs Failed

6 min read
Ollama num_ctx Truncated 287 of 400 Prompts and Never Told Me

Ollama num_ctx Truncated 287 of 400 Prompts and Never Told Me

6 min read
bge-reranker Truncates at 512 Tokens: 31% of My Chunks Got Cut

bge-reranker Truncates at 512 Tokens: 31% of My Chunks Got Cut

1
4
7 min read
LLM-as-a-Judge: Claude Gave 68% of Answers a 7 or 8 Out of 10

LLM-as-a-Judge: Claude Gave 68% of Answers a 7 or 8 Out of 10

1
3
6 min read
GitHub Actions GITHUB_TOKEN Won't Trigger Workflows: 0 of 11 Releases

GitHub Actions GITHUB_TOKEN Won't Trigger Workflows: 0 of 11 Releases

6 min read
tiktoken Undercounts Claude Tokens: 37 of 418 Requests Blew the Limit

tiktoken Undercounts Claude Tokens: 37 of 418 Requests Blew the Limit

7 min read
Speculative Decoding Made My vLLM Server Slower: The Acceptance Math

Speculative Decoding Made My vLLM Server Slower: The Acceptance Math

1
1
7 min read
Anthropic SDK max_retries Stacked on My Retries: 36 Calls per Job

Anthropic SDK max_retries Stacked on My Retries: 36 Calls per Job

1
6 min read
Claude Code Deny Rules Blocked Read(.env). Bash Read It Anyway

Claude Code Deny Rules Blocked Read(.env). Bash Read It Anyway

2
1
6 min read
Claude Code in launchd: My Nightly claude -p Job Worked 6 of 14 Nights

Claude Code in launchd: My Nightly claude -p Job Worked 6 of 14 Nights

1
6 min read
pgvector HNSW Filtering: My RAG Asked for 10 Rows and Got 0

pgvector HNSW Filtering: My RAG Asked for 10 Rows and Got 0

2
6 min read
Nginx proxy_buffering Broke My LLM SSE Stream: 0.4s Became 3.4s

Nginx proxy_buffering Broke My LLM SSE Stream: 0.4s Became 3.4s

1
6 min read
GitHub Actions Cron Was Late 30 Nights Straight, Then Skipped One

GitHub Actions Cron Was Late 30 Nights Straight, Then Skipped One

1
1
7 min read
Claude Parallel Tool Use: My Agent Loop Crashed on 388 of 1,200 Runs

Claude Parallel Tool Use: My Agent Loop Crashed on 388 of 1,200 Runs

4
3
7 min read
Lost in the Middle: My RAG Found the Right Chunk and the LLM Ignored It

Lost in the Middle: My RAG Found the Right Chunk and the LLM Ignored It

7 min read
Claude Prompt Caching Hit Rate Was 0%. One Timestamp Did It

Claude Prompt Caching Hit Rate Was 0%. One Timestamp Did It

7 min read
GitHub Actions Cache Never Saves on a Key Hit: 23 Days of Stale CI

GitHub Actions Cache Never Saves on a Key Hit: 23 Days of Stale CI

2
2
6 min read
MCP stdio Server: One print() Broke 19% of My Tool Calls

MCP stdio Server: One print() Broke 19% of My Tool Calls

1
1
6 min read
loading...