30 seconds about me: I am Liu Gong, an AI Agent engineer with 5 years in AI and computer-vision inspection, fluent in LangChain. Every day I filter the genuinely important AI developments from global sources and explain, in plain language, what happened and what it means. No jargon-stacking, no chasing hype — just judgment you can act on.
Who I am
Hi, I’m Liu Gong — an AI Agent engineer with 5 years in AI and computer-vision inspection. My daily work involves LangChain, agent workflows and LLM deployment, alongside industrial-vision algorithms and real-world rollout.
This site does one simple thing: translate the signals that actually matter in the global AI world into judgments you can understand. Content is published in both English and Chinese for readers worldwide.
Why I built this site
After 5 years in technology, I noticed an awkward gap: much of the AI coverage available in Chinese is either clickbait repackaging or stiff translation. The people who genuinely want to know “what does this mean for me” end up with no answer.
Agents and computer vision in particular are the easiest to get carried away by hype — and I happen to spend every day working hands-on in both. So I built Wenfeng Radar: I filter from global launches, funding rounds, regulation and big-tech moves, keep only the handful of signals that actually change the industry variables, and break them down for you. I’m not an observer — I’m a practitioner in this industry. That’s the difference between me and a pure aggregation feed.
What I believe (principles)
- No manufactured panic: AI disruption is real, but selling anxiety solves nothing. Giving people a framework to judge is what has value.
- No chasing hype: A single headline is noise. Signals that string together into a trend are worth your time.
- Capability ranking always comes third: First ask whether it can be reproduced, then whether it fits your business, and only last how powerful it is.
- Deployment beats showing off: A solution that works in production, closes the loop and saves cost is a good solution.
What I’m watching
- LLMs and inference cost: Model capability is the surface; unit cost is the substance.
- Agents and real deployment: Every step of agents moving from conversation to closed-loop — this is my home turf.
- Computer vision and industrial AI: The real trade-offs among traditional CV, YOLO and LLMs on the production line — I have 5 years of frontline experience.
- Regulation, jobs and industry impact: How rules and people get repriced as AI enters social infrastructure.
What you get here
- Daily picks: Not an information flood — the 1-2 items genuinely worth reading that day.
- Liu Gong’s take: Data-backed personal judgment at the end of every piece, from first-hand observation in real development work.
- Weekly review: 10 key signals of the week strung into one direction you can judge against.
A reminder
What I write is observation and judgment, not investment advice. The AI world moves fast, and my conclusions will iterate — feel free to reach out in the comments or by email. That’s where judgment gets updated.
Contact Liu Gong
For feedback, corrections, topic requests or partnerships, email wenfengleida@wenfengleida.cn. Replies are usually sent within 24–48 hours on weekdays. Chinese-speaking readers can also reach me on the WeChat official account (Wenfeng Radar AI Frontline).
Subscribe for updates
Don’t want to miss the daily digest? Pick one:
- RSS: How to subscribe via RSS, or use https://wenfengleida.cn/feed/ directly
- Email: wenfengleida@wenfengleida.cn