#Mark Chen

123 related articles

Jul 26, 2026·2 min

GPT-5.6 Goes Live with Dual Quota Reset as AI Models Update Rapidly

GPT-5.6 fully launches with enhanced coding, computer operation, and long-horizon agent tasks, plus a dual quota reset. Meanwhile, ByteDance opens its C-Dance 2.5 API and Mistral debuts a single-RGB-camera natural language navigation model.

Jul 25, 2026·5 min

Inside Liang Wenfeng's Closed-Door Meeting: DeepSeek's Philosophy of Restraint and Its AGI Strategy

A full breakdown of DeepSeek founder Liang Wenfeng's 4-hour closed-door meeting: no KPIs, only reasonable profits, open-sourcing top models, and tackling AGI via continual learning—a rare AI strategy confession showing how restraint becomes a core edge.

Jul 24, 2026·4 min

Making $3,000/Month Freelancing with Codex? A Rational Look at AI Coding Side Hustles

Can AI coding tools really earn you $3K/month with zero experience? We break down the marketing hype around Codex freelancing, examine the 4-week roadmap, and reveal what AI-assisted coding can realistically offer.

Jul 23, 2026·3 min

Qwen3.8 Deep Dive: How Does a 2.4 Trillion Parameter MoE Model Claim Second Place?

Alibaba's Qwen3.8 challenges larger models with a 2.4T-parameter MoE architecture, claiming second only to Gemini. A deep dive into MoE mechanics, continuous updates, two-speed release strategy, and real local deployment requirements.

Jul 23, 2026·7 min

Alibaba Open-Sources 2.4T Qwen Model as Chinese AI Sweeps the Global Competitive Landscape

Alibaba open-sources a 2.4 trillion parameter Qwen model and launches the Qwen Token Plan. Chinese models surge, Kimi K3 tops global rankings, and China's AI is reshaping the global competitive landscape.

Jul 22, 2026·6 min

The Complete Codex Guide: From Beginner to Pro with the All-in-One AI Assistant

OpenAI Codex is more than a coding tool—it's an AI agent that writes code, makes slides, and operates browsers. This guide covers its four versions, permission modes, plugins, automation, and six best practices.

Jul 22, 2026·2 min

Cozy AI Kitchen Wraps Up: Microsoft's Heartfelt AI Education Experiment and What It Teaches Us About Content Creation

Microsoft's AI education show Cozy AI Kitchen wraps up. A look at how its 'cozy' style, practical effects, and human focus made AI accessible to everyone.

Jul 21, 2026·2 min

Gradient-Free Evolutionary Model GENREG: How Radial Space Search Bypasses Backpropagation

GENREG-Radial Space is a gradient-free evolutionary optimization model that replaces backpropagation with structured radial space search. This article analyzes its core mechanisms, temporal evolution design, exploration-exploitation balance, and potential as a hybrid paradigm.

Jul 20, 2026·2 min

Husband-Wife Cereal Brand Uses OpenAI Codex to Build Enterprise-Grade Operations System

A founder couple used OpenAI Codex and GPT-5.6 to build an enterprise-grade ops dashboard for cereal brand Three Wishes — no technical co-founder needed.

Jul 20, 2026·2 min

How Designers Can Thrive in the AI Era: Build Fun, Hands-On Agents

AI agents are reshaping what designers do. A senior designer shares why understanding agent infrastructure is essential and how to learn by building fun projects.

Jul 19, 2026·3 min

Math Learning Path for Machine Learning: From Resource Overload to Systematic Mastery

Overwhelmed by ML math courses? This guide maps out linear algebra, calculus, and probability into a practical learning path — from core courses to reference books.

Jul 19, 2026·2 min

GPT-5.6 Launches With a Warning: Compute Capacity Is the Real Test for Large Language Models

Altman warned of possible GPT-5.6 service disruptions at launch, highlighting compute capacity as the true bottleneck for LLMs. Here's what it means for users.

Jul 17, 2026·2 min

Flowith AI Canvas + Codex: Zero-Barrier One-Click AI Short Drama Production Pipeline

Flowith AI canvas + Codex automates AI short drama production end-to-end — from script and characters to scenes. Asset library solves consistency; point-and-edit enables precision control.

Jul 17, 2026·3 min

Waku Agent Deep Dive: The Four Core Pillars of a Local-First AI Assistant

A deep dive into Waku Agent's four pillars: Loop Engineering, three-tier Memory system, Eval assessment, and the Harness scaffold. Full walkthrough of a local-first AI assistant from task execution to memory consolidation.

Jul 16, 2026·2 min

Introduction to AI Agents: Core Concepts, Principles, and Real-World Examples

A deep dive into AI Agent concepts, working principles, and real products. Covers OpenAI Deep Research, Manus, Chensi, the perceive–decide–act loop, and core agent engineering logic.

Jul 15, 2026·4 min

AI Text Detection and Privacy Protection: The Dual Dilemma of LLM Security

How does watermarking work — and why won't companies deploy it? How does differential privacy defend against membership inference attacks? Based on talks by IISc and IIT scholars, this article unpacks the core mechanisms and real challenges in LLM security.

Jul 15, 2026·3 min

Vibe Coding in Practice: A Comprehensive Review and Selection Guide for 6 AI Coding Tools

In-depth comparison of 6 Vibe Coding tools — Claude Code, Trae, Zhipu, Cursor, and more — covering setup cost, Chinese support, pricing, and code quality.

Is LLM the Wrong Foundation for Robot …

Jul 14, 2026·3 min

Is LLM the Wrong Foundation for Robot Intelligence? The Great Debate in Embodied AI

Robotics researcher Ranjay Krishna challenges LLMs as the foundation for robot intelligence. Is language an unnecessary layer between perception and action? A deep dive into VLA models vs. end-to-end architectures.

Jul 14, 2026·5 min

Between Faith and Doubt: A Deep Philosophical Dialogue on the Existence of God

A deep dialogue on God, the nature of faith, and the politicization of religion: Is faith "bad science"? Can divine experience be proven? Exploring the gray area between faith, doubt, and cynicism.

Jul 13, 2026·3 min

Training a 100M-Parameter Model for Under $50: A Complete Record of Building SmoLLM from Scratch

An ML engineer trained SmoLLM, a 109M-parameter LLaMA-style model from scratch for under $50. Full breakdown of architecture, training pitfalls, instruction tuning, and real-world performance.