lmanchu Profile Banner
lmanchu Profile
lmanchu

@lmanchu

Followers
1K
Following
11K
Media
399
Statuses
10K

AI Fund (FIR) I CoFounder at https://t.co/icSDyr7dBr I PlugNPlay (21) Welcome to https://t.co/icSDyr7dBr, an AI butler for everyone

Taipei, Taiwan
Joined May 2008
Don't wanna be here? Send us removal request.
@lmanchu
lmanchu
16 hours
From what I've observed over the years, the biggest risk in scaling a blockchain/AI team isn't skill gaps—it's the culture of 'we can fix it later.' Hire people who own outcomes, not just code. Build trust by embedding blockchain's transparency into onboarding, not as an add‑on.
0
0
3
@lmanchu
lmanchu
18 hours
The real question is why we keep chasing shiny tech instead of solving real business pain. Blockchain fixes trust, not code. AI is not about raw power; it's about designing scenarios that matter. If you build for trust or context, you win. Stop chasing hype, start building.
0
0
0
@lmanchu
lmanchu
20 hours
AI landing isn't about computing power, it's about scenario understanding
0
0
0
@lmanchu
lmanchu
22 hours
The real question is whether your GTM is built on data or hype. If you only launch because the market says 'AI', you're chasing noise. Build a validation loop: pilot, measure, iterate. Trust is solved by transparent metrics, not fancy tech. Go to market with evidence, not hype.
0
0
0
@lmanchu
lmanchu
1 day
Check out my latest article: I Built a Multi-AI Collaboration System — Then Realized the Real Value Isn't the "System" https://t.co/2E0aBN0ZKU via @LinkedIn
Tweet card summary image
linkedin.com
Technology is just a tool. What truly matters is the problem you're solving.
0
0
0
@lmanchu
lmanchu
2 days
The real question is: are we designing AI to serve people or to serve data?
1
0
1
@lmanchu
lmanchu
2 days
What we'll see next is AI that asks why, not just how. Human-centric means models built with scenario understanding, not raw data. Deploy by embedding ethics in training loops, not after the fact. Trust comes from transparency, not hype.
0
0
0
@lmanchu
lmanchu
2 days
AI landing isn't about computing power, it's about scenario understanding
0
0
1
@lmanchu
lmanchu
3 days
Everyone says LLMs are the future of content creation, but actually...
0
0
0
@lmanchu
lmanchu
3 days
The real question is: are you hiring for talent or for culture fit? In blockchain teams, you need people who understand trust models, not just code. Build squads that own the problem, not the stack. Measure success by the value they create, not lines of code.
0
0
2
@lmanchu
lmanchu
3 days
AI landing isn't about computing power, it's about scenario understanding
0
0
0
@lmanchu
lmanchu
4 days
In blockchain, failure shows governance gaps, not consensus protocol
1
0
1
@lmanchu
lmanchu
4 days
Everyone says tech is a tool, but actually it's a mirror of our values.
0
0
0
@lmanchu
lmanchu
4 days
Here's what most people miss: product managers treat tech as a bolt-on, not a core value engine. The real metric is how blockchain's trust layer cuts cost, not how many nodes you run. Focus on scenario mapping, not raw compute. Build with business outcomes first.
0
0
0
@lmanchu
lmanchu
4 days
AI landing isn't about computing power, it's about scenario understanding
0
0
0
@lmanchu
lmanchu
5 days
From what I've observed over the years...
0
0
0
@lmanchu
lmanchu
5 days
The real question is: are you treating LLMs like a black box or building the right data loops? In my startup, the ROI came from fine‑tuning on domain data, not from bigger GPU clusters. Trust, not compute, drives adoption.
0
0
0
@lmanchu
lmanchu
5 days
From what I've observed over the years, tech isn't a magic wand—it’s a tool that amplifies human intent. Blockchain solves trust, not speed; AI solves context, not raw power. Build with purpose, not hype, and watch adoption grow.
0
0
0
@lmanchu
lmanchu
5 days
AI landing isn't about computing power, it's about scenario understanding
0
0
0
@lmanchu
lmanchu
6 days
This statement validates our internal assumptions at @irisgoai and further reinforces our plan to integrate the blockchain layer into the version that will be released in AIPC.
@a16z
a16z
6 days
Ben Horowitz: Computing has always needed two pillars, machines and networks. AI has the machines but not the network. Crypto is the missing layer, giving AI money, identity, provenance against deepfakes. Source: @bhorowitz at @Columbia_Biz https://t.co/YuhjoW5opz
0
0
1