yuer

Yuer — Independent AI Systems Architect Building the Expression-Driven Cognitive Architecture (EDCA OS): a deterministic, auditable execution layer for LLMs. Focused on: deterministic RAG & reproduc

Joined on Nov 20, 2025 lipxtk@gmail.com

yuer

Jan 5

Why LLMs Break in Production (and Why It’s Not a Model Problem)

#llm #softwareengineering #systemdesign

3 min read

Want to connect with yuer?

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

Create Account

Already have an account? Sign in

yuer

Jan 4

Controllable AI Must Sit in the Control Plane — Otherwise It Shouldn’t Exist

#ai #architecture #systemdesign

2 min read

yuer

Jan 3

When Community AI Breaks, It’s Rarely the Model

#ai #community #data #systemdesign

3 min read

yuer

Jan 2

A Fail-Closed Gate for Rust AI Assistants

#discuss #codequality #rust #ai

2 min read

yuer

Jan 1

Controllable AI: A Runtime Legitimacy Layer for AI Governance

#architecture #agents #llm #ai

3 min read

yuer

Jan 1

Stop Giving AI the Steering Wheel

#architecture #agents #security #ai

2 min read

yuer

Dec 31 '25

Your AI Agent Isn’t Unsafe — It’s Unstoppable (And That’s the Problem)

2 min read

yuer

Dec 30 '25

AI Systems That Can’t Say “No” Are Not Production-Ready

#ai #security #systemdesign

2 min read

yuer

Dec 30 '25

Why AI Agents Break in Production (And Why It’s Not a Prompt Problem)

#agents #ai #architecture

5 min read

yuer

Dec 27 '25

Why Interruptible Voice AI Is a Systems Problem (Not an AI Problem)

#discuss #systemdesign #architecture #ai

2 min read

yuer

Dec 27 '25

Why Distributed Query Engines Always Accumulate Complexity in the Execution Layer

#architecture #database #rust

3 min read

yuer

Dec 26 '25

Starting a Distributed Query Engine with a Minimal Rust Execution POC

#rust

3 min read

yuer

Dec 25 '25

Why Post-Hoc Moderation Fails in Real-Time Systems

#architecture #security #systemdesign

4 min read

yuer

Dec 25 '25

I Stopped Automating ecisions with AI — and Designed a Better Collaboration Instead

#discuss #design #automation #ai

4 min read

yuer

Dec 24 '25

Prompt Engineering Is Fragile — Human–AI Collaboration Is the Real Interface

#discuss #llm #promptengineering #ai

2 min read

yuer

Dec 23 '25

Powerful LLMs Are Not the Problem — Using Them “Raw” Is A systems-engineering view for builders

#softwareengineering #llm #ai #architecture

5 min read

yuer

Dec 23 '25

The GPT Client Is Becoming the Runtime for Human–AI Collaboration

#ai #architecture #llm

3 min read

yuer

Dec 21 '25

AI decision systems break without strategy stability

#ai #architecture #llm

2 min read

yuer

Dec 21 '25

Using a Single GPT Client as a Language Runtime (No API, No Agents)

#chatgpt #llm #architecture #ai

2 min read

yuer

Dec 20 '25

Controllable AI: Why Enterprise AI Fails at Behavior, Not Models

#ai #machinelearning #architecture #enterprisetech

6 min read

yuer

Dec 19 '25

AI Is Powerful — But Most of Us Are Still Using It the Wrong Way

2 min read

yuer

Dec 18 '25

If the Same Input Gives Different Results, It’s Not a Decision System

#discuss #ai #systemdesign

2 min read

yuer

Dec 17 '25

Using GPT as a Code Auditor (Not a Code Generator)

#ai #security #testing

2 min read

yuer

Dec 15 '25

Chat-based AI Breaks Down for Long-Running Work

#discuss #llm #ai #productivity

2 min read

yuer

Dec 9 '25

You Don’t Need RAG to Build a Domain Sentiment Engine — A Single LLM + JSON Works

#ai #machinelearning #llm #nlp

2 min read

yuer

Dec 9 '25

A RAG-Free Technique That Makes LLM Outputs Stable, Predictable, and Auditable

#rag #llm #architecture #ai

2 min read

yuer

Dec 8 '25

I Turned ChatGPT Web Into a Controllable Risk Engine — No Code, No API, No RAG

3 min read

yuer

Dec 7 '25

Turning ChatGPT into a Deterministic Flight-Risk Runtime (FRR Demo + GitHub Repo)

2 min read

yuer

Nov 20 '25

Why RAG and Agent Systems Are Unstable — A Minimal Deterministic Planner POC

#agents #rag #architecture #llm

2 min read

yuer

Nov 20 '25

Fixing Identity Drift in AI Image Generation with a Deterministic Constraint Layer (Minimal PoC Inside)

#algorithms #ai #deeplearning #machinelearning

2 min read

yuer

Nov 20 '25

Deterministic RAG: A Drop-in Replacement for GraphRAG’s Unstable Planning

#architecture #llm #rag

3 min read

DEV Community

yuer

Badges

Writing Debut

Why LLMs Break in Production (and Why It’s Not a Model Problem)

Want to connect with yuer?

Controllable AI Must Sit in the Control Plane — Otherwise It Shouldn’t Exist

When Community AI Breaks, It’s Rarely the Model

A Fail-Closed Gate for Rust AI Assistants

Controllable AI: A Runtime Legitimacy Layer for AI Governance

Stop Giving AI the Steering Wheel

Your AI Agent Isn’t Unsafe — It’s Unstoppable (And That’s the Problem)

AI Systems That Can’t Say “No” Are Not Production-Ready

Why AI Agents Break in Production (And Why It’s Not a Prompt Problem)

Why Interruptible Voice AI Is a Systems Problem (Not an AI Problem)

Why Distributed Query Engines Always Accumulate Complexity in the Execution Layer

Starting a Distributed Query Engine with a Minimal Rust Execution POC

Why Post-Hoc Moderation Fails in Real-Time Systems

I Stopped Automating ecisions with AI — and Designed a Better Collaboration Instead

Prompt Engineering Is Fragile — Human–AI Collaboration Is the Real Interface

Powerful LLMs Are Not the Problem — Using Them “Raw” Is A systems-engineering view for builders

The GPT Client Is Becoming the Runtime for Human–AI Collaboration

AI decision systems break without strategy stability

Using a Single GPT Client as a Language Runtime (No API, No Agents)

Controllable AI: Why Enterprise AI Fails at Behavior, Not Models

AI Is Powerful — But Most of Us Are Still Using It the Wrong Way

If the Same Input Gives Different Results, It’s Not a Decision System

Using GPT as a Code Auditor (Not a Code Generator)

Chat-based AI Breaks Down for Long-Running Work

You Don’t Need RAG to Build a Domain Sentiment Engine — A Single LLM + JSON Works

A RAG-Free Technique That Makes LLM Outputs Stable, Predictable, and Auditable

I Turned ChatGPT Web Into a Controllable Risk Engine — No Code, No API, No RAG

Turning ChatGPT into a Deterministic Flight-Risk Runtime (FRR Demo + GitHub Repo)

Why RAG and Agent Systems Are Unstable — A Minimal Deterministic Planner POC

Fixing Identity Drift in AI Image Generation with a Deterministic Constraint Layer (Minimal PoC Inside)

Deterministic RAG: A Drop-in Replacement for GraphRAG’s Unstable Planning