Jacky Lin (@renlas_bern) header
Jacky Lin (@renlas_bern) profile picture

Building AI workflows that run B2B outreach end-to-end Founder, Wing Future · @aitobrands

Guangzhou, China
Joined April 2011
Photo by Jacky Lin (@renlas_bern)Photo by Jacky Lin (@renlas_bern)Photo by Jacky Lin (@renlas_bern)Photo by Jacky Lin (@renlas_bern)Photo by Jacky Lin (@renlas_bern)Photo by Jacky Lin (@renlas_bern)Photo by Jacky Lin (@renlas_bern)Photo by Jacky Lin (@renlas_bern)Photo by Jacky Lin (@renlas_bern)Photo by Jacky Lin (@renlas_bern)
Muse on my Apple Watch — and somehow, even when it’s working, it looks this charming. We’re at the point where AI agents can live on the devices we already wear. A short conversation is enough to put them to work, anytime, anywhere.
14
Anyone who has a business" isn’t an ICP. It’s a wish list. Write three hard reject rules first, then rewrite the email. Outreach usually dies at Find, not the subject line
PROSPECT: "Cold email doesn't work for my agency. I've tried it." ME: "What's in the email?" PROSPECT: "Everything. SEO, ads, social, web design, automation, lead gen." ME: "Who's it going to?" PROSPECT: "Anyone who has a business." ME: "Which of those services do they want?" PROSPECT: "No idea, whichever one catches them, I guess." Stop sending ur prospects menus... Pick the one service you're best at, send it to one niche, and see what happens.
43
Your AI sales “system” still needs you in the room to paste the ICP every morning? That’s not a workflow. That’s a fancy chat tab. Here’s the difference most founders miss: Chat-with-AI = you re-explain the buyer, the offer, and the rules every time. It feels productive. It dies the day you’re busy. Workflow Thinking = one step has fixed inputs, a fixed output file, and a review checklist someone else (or tomorrow-you) can run without the story. If you can’t leave the room for 2 hours and still get a usable list out the other side, you don’t have a step yet. Try this for 5 days — don’t rebuild anything: 1. Pick ONE step only: Find, Qualify, or Outreach. 2. Freeze the inputs (ICP doc, signal list, or shortlist — whatever that step eats). 3. Freeze the output format (a sheet/row template, not a paragraph in chat). 4. Freeze the human review checklist (3–5 yes/no questions max). 5. Change only the list going in. Not the prompt. Not the model. Not the “system.” What you should see by day 5: • Same morning ritual, shorter • Output someone else can open without asking you “what did you mean?” • Clear failure modes (bad signal, weak qualify rule, weak reason-to-send) instead of “the AI was off today” What usually breaks first: • Find without signals → louder CRM, same junk • Qualify as an essay → pretty text, no reject rule • Outreach as “personalize this” → first line automation, zero reason to reply Fix the broken step. Don’t stack another tool on top of a chat. A finished step beats a half-built AI sales brain. Save this if you’re stuck “building the system” instead of shipping tomorrow’s list.
22
Most AI outreach starts at Find. Start one step earlier. Write who you will not sell to until a junior hire can reject half a list without pinging you. Find without that filter just makes your CRM louder.
24
Most B2B AI projects don't fail because of the model. They fail because nobody mapped the workflow first. What triggers it?
What context does AI need?
Where does a human review?
How do you measure it? No answers = faster chaos.
1
20
Correct!
There's so many posts on the timeline about the optimal time to do something, or the optimal age, or the optimal conditions necessary, etc. It's all cope. There's is no optimal time. The optimal time is now. Just do what it is you want to do. Do it badly. Do it and suck. Just do it. Nobody actually cares. It's all in your head.
16
network is one of the best thing, you can see what others do and what they get.
If you're not sure what to do with your life, just do something. Write, record, publish, call, email, network. Anything. Momentum is always built through action, not thinking about action.
11
This really resonated with me. I’ve written books and developed AI applications. In both cases, I promoted the product while it was still a work in progress, so that many users could interact with it and learn about it. That’s exactly what you need to do in the AI era!
Two founders launch the same AI app on the same day. Founder A spent 6 months perfecting the product. 0 followers. 0 email list. Launch day: 40 signups, mostly friends. Founder B spent those 6 months posting about what they were building. 8,000 followers. 1,200 people on the waitlist. Launch day: 900 signups. Same idea. Same tools. Different outcome. Founder A's app is better. Nobody will ever find out. In 2026 the code takes a weekend. The audience takes a year. If you're building in silence, you're not being humble. You're just making it harder for anyone to find you. Post the ugly version. Share the numbers. Start now.
5
Most people use AI like a smarter Google. Ask. Get an answer. Feel busy. That’s not a system. A system starts with the workflow: who you want → how you find them → how you qualify → how you reach out. Then AI only executes each step. If you skip the workflow, you’re just chatting.
8
Marketing is also a part of success.
You can spend 5 years, working harder than anyone else, writing the best book in the world, but if nobody knows about it, then you will not be a successful author. This is the downfall of most creatives. They hate marketing, but marketing is 80% of their success.
6
Amazing!
Turns out the future does have flying cars
6
Grok Voice 3.0 in Grok Bot is 🔥 SpaceXSI just launched Grok Voice 3.0, its new speech-to-speech model, which is currently beating GPT Live 1 and Gemini 3.8 Flash Live on all major benchmarks. I tried it in Grok Bot, and the experience is great. It’s fast, responsive, and makes voice interactions feel much better.
81
101
5
1,160
295,672
Most people use AI like a smarter Google. Ask a question. Get an answer. Feel productive. That’s not a system. A system is when you decide the workflow first: who you want → how you find them → how you qualify → how you reach out. Then AI just executes each step. I’m writing about this for export / B2B teams building real sales workflows — not prompt tips. If you’re still “asking AI for emails,” you’re one layer too early.
5
Grok bot is the best AI agent!
99% 的人都不会用 Grok Bot! 官方下场教你如何跑几十个 Grok Bot 组成能自我迭代升级的团队 视频时长 1 个小时,非常干,一点稀的都没有,先码后看👀
16
We've unified plans across Grok, Cursor, and X Premium. Introducing The X Subscription.
1,574
1,930
1,970
17,650
19,581,202
Jacky Lin retweeted
Dusting off this tweet from April because this gap in shared understanding of LLM capability is *widening*. It's now less "two groups of people speaking past each other" and more a sharp funnel. - Napkin math somewhere around 6B people (~75% of the population) have barely come in contact with LLMs at all. - Around 1-2B (~20%) are casual and infrequent users of free-tier ChatGPT-like products. This group sees derpy chatbots and treats them a bit like a better Google search, a writing aid, or etc. Maybe an agent tries to book you a flight. I have non-tech friends who (reasonably, imo) say they have not much use for it at all. Even many professionals outside of math&code are in this tier. For example, execs and many other functions spend a lot of their time talking to other people, so while they understand what is happening intellectually, it is still second-hand and a bit abstract. - Now we get to professional use of frontier-grade LLMs in math&code. Somewhere around 20M people (0.2%) see first-hand that large, complex projects that used to take them weeks/months can now be completed by agents with a prompt. Building apps, copying apps, translating apps, decompiling apps from binaries... This has all happened very quickly and recently - less than 1 year ago, I was writing code manually by hand, typing memorized computer code commands into a code editor character by character, occasionally pressing Tab to autocomplete a little chunk of code. - And finally we get to the ~5,000 people (~0.00006%) with access to frontier-grade systems internally. The external world has seen the preview. It looks like swarms of thousands of agents collaborating over weeks on software mega projects: minting zero days, running cyber attacks and defenses at machine speeds, discovering new science, advancing the frontier of mathematics. Things that would have taken top professionals in the industry years of work. Meanwhile, human review and comprehension are starting to fall behind. For example, people are still involved in the "archeology" of the OpenAI-HF incident from many months ago. Mathematicians may be poring over the 722 manuscripts on frontier mathematics for a while. The funnel is driven by a combination of factors: 1. The impact scales with ambition, problem size, and horizon. A question with a paragraph answer barely stresses the system. You need a reservoir of big, difficult problems that you really care about. This aspect drives the consumer / professional dimension of the funnel. 2. The jaggedness of the system (which I have written about a lot separately). Capability peaks in domains that are digital, verifiable and economically valuable. This is because LLM capability emerges from reinforcement learning on verifiable rewards on a curated environment mixture driven by revenue potential. This aspect primarily drives the area (e.g. math&code) dimension of the funnel. 3. Access. Free-tier, paid-tier, internal. So this is the weirdness of the moment. The general public has mostly not interacted with these systems. When they have, it looks like a derpy chatbot. The majority of professionals still see only a modest uplift. And a small sliver of professionals are experiencing the vertigo of the curve going vertical. And it is all happening at the same time.
Judging by my tl there is a growing gap in understanding of AI capability. The first issue I think is around recency and tier of use. I think a lot of people tried the free tier of ChatGPT somewhere last year and allowed it to inform their views on AI a little too much. This is a group of reactions laughing at various quirks of the models, hallucinations, etc. Yes I also saw the viral videos of OpenAI's Advanced Voice mode fumbling simple queries like "should I drive or walk to the carwash". The thing is that these free and old/deprecated models don't reflect the capability in the latest round of state of the art agentic models of this year, especially OpenAI Codex and Claude Code. But that brings me to the second issue. Even if people paid $200/month to use the state of the art models, a lot of the capabilities are relatively "peaky" in highly technical areas. Typical queries around search, writing, advice, etc. are *not* the domain that has made the most noticeable and dramatic strides in capability. Partly, this is due to the technical details of reinforcement learning and its use of verifiable rewards. But partly, it's also because these use cases are not sufficiently prioritized by the companies in their hillclimbing because they don't lead to as much $$$ value. The goldmines are elsewhere, and the focus comes along. So that brings me to the second group of people, who *both* 1) pay for and use the state of the art frontier agentic models (OpenAI Codex / Claude Code) and 2) do so professionally in technical domains like programming, math and research. This group of people is subject to the highest amount of "AI Psychosis" because the recent improvements in these domains as of this year have been nothing short of staggering. When you hand a computer terminal to one of these models, you can now watch them melt programming problems that you'd normally expect to take days/weeks of work. It's this second group of people that assigns a much greater gravity to the capabilities, their slope, and various cyber-related repercussions. TLDR the people in these two groups are speaking past each other. It really is simultaneously the case that OpenAI's free and I think slightly orphaned (?) "Advanced Voice Mode" will fumble the dumbest questions in your Instagram's reels and *at the same time*, OpenAI's highest-tier and paid Codex model will go off for 1 hour to coherently restructure an entire code base, or find and exploit vulnerabilities in computer systems. This part really works and has made dramatic strides because 2 properties: 1) these domains offer explicit reward functions that are verifiable meaning they are easily amenable to reinforcement learning training (e.g. unit tests passed yes or no, in contrast to writing, which is much harder to explicitly judge), but also 2) they are a lot more valuable in b2b settings, meaning that the biggest fraction of the team is focused on improving them. So here we are.
2,269
5,947
2,300
44,405
7,050,696
How much traffic Google AI Mode is referring to sites? @Similarweb reclassified its traffic from "Gen AI" to "Organic Search" on September 1st, and we can easily see how much Gen AI traffic drops on September 1st for any site. For many of the top sites of the web, also used as citation sources, such as YouTube, Reddit, Facebook, it has been a 60-70% drop in Gen AI traffic (which is still a fraction of organic search). If you have access to Similarweb, go and take a look 😉
8
13
1
71
8,399
When Elon calls for a rap feature, you don't say no. 😎🎤 #AI #AIrap #AIGC
146