ahwai zhang

Independent developer at Sanva. Building apps for personal knowledge and creator workflows.

Joined on Oct 27, 2026 https://www.sanva.top/

ahwai zhang

Nov 3

I Shipped 5 AI Micro-Apps on iOS in 30 Days: The Good, the Bad, and the Stack That Made It Possible

#ai #buildinpublic #ios #startup

2 min read

ahwai zhang

Nov 2

The True Cost of Context Switching: Why Modern Engineering is a Memory Problem, Not a Time Problem

#developers #productivity #softwareengineering

2 min read

ahwai zhang

Nov 1

Cognitive Offloading: Combining FSRS Spaced Repetition with Fluid AI Context Memory

2 min read

ahwai zhang

Oct 31

The Solo Creator Stack: How AI Memory & Rapid Asset Gen Saved 15 Hours on Product Launch Weekend

2 min read

ahwai zhang

Oct 30

On-Device Semantic Retrieval on iOS: How Lumo Solves Note Fatigue Without Cloud Latency

#privacy #ai #mobile #ios

2 min read

ahwai zhang

Oct 30

The Cognitive Architecture of AI Memory: Why Vector Search Outperforms Hierarchical Folders

#search #architecture #ai #database

2 min read

ahwai zhang

Oct 30

Solving AI Agent Amnesia: Building a Self-Hosted Shared Memory Mesh for Coding LLMs

#softwaredevelopment #agents #llm #ai

3 min read

ahwai zhang

Oct 30

Why Ephemeral Sessions Cripple AI Coding Agents (And How to Build a Self-Hosted Long-Term Memory)

#agents #llm #ai #softwareengineering

3 min read

ahwai zhang

Oct 30

On-Device Semantic Retrieval on iOS: How Lumo Solves Note Fatigue Without Cloud Latency

#mobile #privacy #ai #ios

2 min read

ahwai zhang

Oct 28

Introducing Lumo (Memory Copilot): Never Lose Context in Your Daily Tech Workflow

#productivity #tools #ai

1 min read

DEV Community

ahwai zhang

Badges

Writing Debut

I Shipped 5 AI Micro-Apps on iOS in 30 Days: The Good, the Bad, and the Stack That Made It Possible

The True Cost of Context Switching: Why Modern Engineering is a Memory Problem, Not a Time Problem

Cognitive Offloading: Combining FSRS Spaced Repetition with Fluid AI Context Memory

The Solo Creator Stack: How AI Memory & Rapid Asset Gen Saved 15 Hours on Product Launch Weekend

On-Device Semantic Retrieval on iOS: How Lumo Solves Note Fatigue Without Cloud Latency

The Cognitive Architecture of AI Memory: Why Vector Search Outperforms Hierarchical Folders

Solving AI Agent Amnesia: Building a Self-Hosted Shared Memory Mesh for Coding LLMs

Why Ephemeral Sessions Cripple AI Coding Agents (And How to Build a Self-Hosted Long-Term Memory)

On-Device Semantic Retrieval on iOS: How Lumo Solves Note Fatigue Without Cloud Latency

Introducing Lumo (Memory Copilot): Never Lose Context in Your Daily Tech Workflow