Andrew Ng
DeepLearning.AI, AI Fund and AI Aspire
Andrew Ng announces AI tools and educational products to 2.5M engineers, driving 4,246 avg engagements per post by leading with concrete problems his audience faces (outdated APIs, job market shifts, no-code barriers). His distinctive move: he solves the problem in the same post, then shares the open-source tool or course link, collapsing the gap between awareness and action.
3/wk
7 posts in 30d
4.2K
avg per post
5.1%
comments / total
5.6%
reposts / total
problem-solution
Sun · 1 PM CET
Voice & tone
Ng writes like a founder-teacher who has seen the problem firsthand. He's direct without hype, grounded in specifics ("Claude Code calls the older chat completions API instead of the newer responses API"), and optimistic about what's possible. He uses "I" frequently ("I'm excited," "I've just launched," "I'd like to share") to build trust, and he credits collaborators by name, signaling generosity over ego.
Signature phrases
- “I'm excited to announce”
- “Why this matters:”
- “AI won't replace workers, but workers who use AI will replace workers who don't”
- “If you've never written code before, this is for you”
- “This will be the best place to start”
- “See [resource] for more details”
Top 3 posts, dissected
Their highest-engagement posts, broken down line by line. Steal the structure.
I'm excited to announce Context Hub, an open tool that gives your coding agent the up-to-date API documentation it needs. Install it and prompt your agent to use it to fetch curated docs via a simple CLI. (See image.) Why this matters: Coding agents often use outdated APIs and hallucinate parameters. For example, when I ask Claude Code to call OpenAI's GPT-5.2, it uses the older chat completions API instead of the newer responses API, even though the newer one has been out for a year. Context Hub solves this. Context Hub is also designed to get smarter over time. Agents can annotate docs with notes — if your agent discovers a workaround, it can save it and doesn't have to rediscover it next session. Longer term, we're building toward agents sharing what they learn with each other, so the whole community benefits. Thanks Rohit Prsad and Xin Ye for working with me on this! npm install -g @aisuite/chub GitHub: https://lnkd.in/gGeSApnE
"I'm excited to announce Context Hub, an open tool that gives your coding agent the up-to-date API documentation it needs." This works because it leads with founder energy ("I'm excited"), names the tool immediately, and states the benefit in one clause. The reader knows in 12 words whether they care.
Announcement (hook) > Why This Matters (problem + concrete example) > Solution Details (how it works, what makes it smart) > Social Proof (GitHub stars, community contributions) > Installation Command (frictionless next step) > Credits (names collaborators).
- Concrete failure example (Claude Code using older API) makes the problem real, not abstract
- "Why this matters:" subheading breaks up the wall of text and signals importance
- Forward-looking vision (agents sharing learnings with each other) expands the post beyond the tool
- npm install command at the end removes friction; reader can act immediately
Ends with installation command and GitHub link. No soft CTA. The action is embedded in the post itself, so engagement comes from people installing, starring, and commenting on their results.
Lead with the tool name and benefit in one sentence, then immediately show why it matters with a specific failure case your audience recognizes.
Job seekers in the U.S. and many other nations face a tough environment. At the same time, fears of AI-caused job loss have — so far — been overblown. However, the demand for AI skills is starting to cause shifts in the job market. I’d like to share what I’m seeing on the ground.…
"Job seekers in the U.S. and many other nations face a tough environment. At the same time, fears of AI-caused job loss have — so far — been overblown." This hook works by validating the reader's anxiety (tough job market) then immediately offering relief (fears overblown), creating cognitive tension that demands resolution.
Problem Statement (job market is hard) > Counterintuitive Claim (AI job loss fears overblown) > Nuance (but shifts are happening) > Ground Truth (layoffs were pandemic corrections, not AI automation) > Specific Risk Groups (call-center operators, translators, voice actors) > Actionable Reframe (workers who use AI will replace those who don't).
- Three-part opening (bad news, good news, nuance) creates narrative momentum
- "The reality is AI just doesn't work that well yet" is a disarming admission that builds credibility
- Specific job categories at risk (translators, voice actors) make the threat concrete without being alarmist
- The phrase "AI won't replace workers, but workers who use AI will replace workers who don't" is a quotable reframe that shifts agency to the reader
Post cuts off mid-sentence ("For instance, beca...") which forces readers to click through to LinkedIn to read the full post, driving engagement metrics.
Use a three-part hook (problem, counterintuitive claim, nuance) to create tension, then resolve it with a specific, quotable insight that readers will want to share.
If you’ve never written code before, this is for you. I’ve just launched a course that shows you, in less than 30 minutes, how to describe an idea for an app and build it with AI. In this course, you'll build a working web application - a funny interactive birthday message gener…
"If you've never written code before, this is for you." This hook works by explicitly including a reader who might otherwise scroll past. It removes the self-doubt barrier in one sentence and immediately follows with a concrete outcome (build a working app in under 30 minutes).
Direct Address (if you've never coded, this is for you) > Outcome + Timeframe (build a web app in less than 30 minutes) > Concrete Project (birthday message generator) > Customization Process (tell AI how you want it changed) > Skills Gained (bulleted list of three outcomes) > Vendor Neutrality (works with ChatGPT, Gemini, Claude) > Closing Encouragement (you'll be surprised at what you can build).
- Direct address ("this is for you") removes gatekeeping and invites the excluded audience
- Concrete project (birthday message generator) is specific enough to visualize but simple enough to feel achievable
- Bulleted skills list breaks up text and makes the value scannable
- Vendor neutrality ("we're vendor neutral") removes friction for readers locked into a specific AI tool
- Closing line ("I think you will be surprised at what you can build") is warm and encouraging, not salesy
Ends with a soft CTA ("Please join me and take your first step!") and an acknowledgment of experienced engineers ("if you're already an experienced engineer, plea...") which signals inclusivity and removes the objection that the course is too basic.
When teaching a new skill to beginners, lead with direct address, show a concrete project they'll build, list the skills they'll gain, and close with encouragement rather than urgency.
What drives their engagement
Comments spike when Ng asks a direct question ("Should there be a Stack Overflow for AI coding agents?") or when he shares a contrarian take that readers want to debate ("AI won't replace workers, but workers who use AI will replace workers who don't"). The conversation index is low (5.1%), so comments are sparse but substantive.
Reposts peak on posts with quotable insights, specific tools readers can install immediately, or announcements that signal industry shifts (Coursera-Udemy merger). The 5.6% repost rate suggests his audience shares to signal expertise and insider knowledge.
Posts with concrete resources (GitHub links, course URLs, installation commands) and skill-building content (no-code education) likely drive saves. Posts that reframe anxiety (job market reality checks) are also savable because readers return to them when facing decisions.
Content pillars
AI Agent Tools & Infrastructure
28%Launches open-source tools (Context Hub) that solve specific agent problems like outdated API documentation, with installation commands and GitHub links.
No-Code/Low-Code Education
23%Teaches non-engineers how to build with AI in under 30 minutes, removing the coding barrier and positioning AI as a creative tool.
AI Job Market Reality Check
18%Counters hype with ground truth about layoffs, automation risk, and the real skill shift happening in tech hiring.
Platform & Partnership Announcements
15%Shares major business news (Coursera-Udemy merger) with personal context on why it matters to learners.
Community-Driven Agent Development
16%Explores how agents can learn from each other and share knowledge, positioning the future of AI collaboration.
Hook patterns
Excited Announcement + Immediate Problem
I'm excited to announce [tool]. [Specific problem it solves with concrete example].
“I'm excited to announce Context Hub, an open tool that gives your coding agent the up-to-date API documentation it needs. Install it and prompt your agent to use it to fetch curated docs via a simple CLI. (See image.) Why this matters: Coding agents often use outdated APIs and hallucinate parameters.”
→ Combines founder energy with immediate relevance; the reader knows within 15 words if this is for them.
Direct Address to Excluded Group
If you've never [done X], this is for you. [Outcome in timeframe].
“If you've never written code before, this is for you. I've just launched a course that shows you, in less than 30 minutes, how to describe an idea for an app and build it with AI.”
→ Removes self-doubt by explicitly including the reader; the 30-minute promise is specific enough to feel achievable.
Problem Statement + Market Context
Job seekers [face condition]. At the same time, [counterintuitive claim]. However, [real shift happening].
“Job seekers in the U.S. and many other nations face a tough environment. At the same time, fears of AI-caused job loss have — so far — been overblown. However, the demand for AI skills is starting to cause shifts in the job market.”
→ Sets up tension (bad news, then good news, then nuance) that makes readers want the full story.
Direct Question + Context Update
Should there be [provocative question]? [Recent update with metrics]. [Call to action].
“Should there be a Stack Overflow for AI coding agents to share learnings with each other? Last week I announced Context Hub (chub), an open CLI tool that gives coding agents up-to-date API documentation. Since then, our GitHub repo has gained over 6K stars.”
→ The question creates curiosity; the metrics prove momentum; readers feel like they're part of a movement.
Bold Statement + Credibility Signal
[Announcement]. [Company I've admired] has done [specific work]. I'm excited about [outcome].
“Coursera has entered into a definitive agreement to combine with Udemy. Udemy -- a company I've long admired -- has done tremendous work upskilling millions. I'm excited about the combination's ability to serve learners.”
→ The personal endorsement ("I've long admired") converts a press release into a trusted recommendation.
Contrarian Reframe + Actionable Truth
[Common fear]. But [reality check]. Instead, [what's actually happening].
“fears of AI-caused job loss have — so far — been overblown. However, the demand for AI skills is starting to cause shifts in the job market. Instead, a common refrain applies: AI won't replace workers, but workers who use AI will replace workers who don't.”
→ Defuses anxiety while giving readers a clear survival strategy; the phrase is quotable and shareable.
Top-performing hooks
“I'm excited to announce Context Hub, an open tool that gives your coding agent the up-to-date API documentation it needs. Install it and prompt your agent to use it to fetch curated docs via a simple CLI. (See image.)”
→ The post clearly identifies a common problem faced by developers using AI coding agents (outdated APIs) and presents a solution (Context Hub). It also highlights the collaborative and community-driven
“Job seekers in the U.S. and many other nations face a tough environment. At the same time, fears of AI-caused job loss have — so far — been overblown. However, the demand for AI skills is starting to cause shifts in the job market. I’d like to share ”
→ The post addresses a timely and relevant concern (AI job displacement) with a nuanced perspective, offering both reassurance and actionable insights. The inclusion of specific examples and future pred
“If you’ve never written code before, this is for you. I’ve just launched a course that shows you, in less than 30 minutes, how to describe an idea for an app and build it with AI.”
→ The post directly addresses a common pain point (lack of coding skills) and offers a quick, easy solution (AI-powered app building). The promise of building a working app in 30 minutes and vendor neut
“Should there be a Stack Overflow for AI coding agents to share learnings with each other?”
→ The post opens with a direct question to encourage immediate engagement. It then provides a compelling update on a successful open-source project, fostering a sense of community and inviting further p
“Coursera has entered into a definitive agreement to combine with Udemy.”
→ The post announces a significant industry event, immediately establishing authority. The clear and concise language, coupled with the mention of leadership roles, reinforces credibility and generates
“Will AI create new job opportunities? My daughter Nova loves cats, and her favorite color is yellow. For her 7th birthday, we got a cat-themed cake in yellow by first using Gemini’s Nano Banana to design it, and then asking a baker to create it using”
→ The post uses a relatable personal anecdote to introduce a broader discussion about AI's impact on jobs. It balances concerns about job displacement with optimistic predictions about new opportunities
“I'm delighted that Coursera and Udemy have come together as one company to serve learners.”
→ The post leverages a major industry announcement combined with thought leadership on AI's impact on work, positioning the author as a visionary leader while emphasizing the mission-driven purpose behi
“New course: Document AI: From OCR to Agentic Doc Extraction, built with LandingAI, where I'm executive chairman, and taught by David Park and Andrea Kropp.”
→ The post clearly identifies a common problem (data locked in documents) and offers a specific solution (LandingAI's ADE framework) with tangible skills gained. The mention of practical applications li
“Apple just named its latest laptop Neo -- same name as my son! Should I buy one?”
→ The post leverages a personal connection (son's name) to a popular product, creating an relatable and engaging question for the audience. It also hints at a potential tech demo, further sparking inter
“Important new course: Agent Skills with Anthropic, built with Anthropic and taught by Elie Schoppik!”
→ The post announces a valuable new course from a credible source (Anthropic and Elie Schoppik), offering practical skills in a rapidly growing field. The clear structure and specific learning outcomes
What works
- Problem Statement hooks outperform all others (6 posts, 6,103 avg engagement). Ng leads with a specific failure case ("Claude Code uses the older chat completions API") rather than a generic problem, making the relevance immediate.
- Direct Address to excluded audiences ("If you've never written code before") generates 9,593 engagements. This removes self-doubt and invites readers who might otherwise skip the post.
- Concrete installation commands and GitHub links at the end drive action. Post 1 ends with "npm install -g @aisuite/chub" and GitHub URL, turning engagement into adoption.
- Quotable reframes ("AI won't replace workers, but workers who use AI will replace workers who don't") generate 10,561 engagements. These are short enough to tweet, specific enough to be credible, and actionable enough to guide behavior.
- Credibility signals through collaboration credits ("Thanks Rohit Prsad and Xin Ye") and personal endorsements ("Udemy -- a company I've long admired") build trust without sounding promotional.
- Three-part hooks (problem, counterintuitive claim, nuance) create narrative tension that demands resolution. Post 2 uses this pattern to convert anxiety into engagement.
Steal this for your own LinkedIn
- Start your next post with a direct address to an excluded audience ("If you've never [done X], this is for you") rather than a generic problem statement. Test this on your next educational post.
- End announcement posts with a concrete next step (installation command, course link, GitHub URL) rather than a soft CTA. Remove friction between awareness and action.
- Use a three-part opening hook: validate the reader's pain, offer a counterintuitive claim, then add nuance. This pattern appears in Ng's highest-engagement posts and creates cognitive tension.
- Include a quotable reframe that shifts agency to the reader ("Workers who use AI will replace workers who don't"). Make it short enough to copy-paste and specific enough to guide behavior.
- Credit collaborators by name and explain why you admire them. This builds trust and signals generosity, not ego.
- Post on Sunday at noon UTC if your audience is global professionals. Ng's data shows this is his best window.
Compare Andrew with…
vs Eric Partaker
Andrew Ng vs Eric Partaker: Reach versus Conversation on LinkedIn
vs Reid Hoffman
Andrew Ng vs Reid Hoffman: Volume beats depth on LinkedIn
vs Allie K. Miller
Allie K. Miller vs Andrew Ng: Insider access beats product launches
vs Steven Bartlett
Andrew Ng vs Steven Bartlett: Problem-solving versus storytelling at scale
Get the full Andrew Ng playbook
Hook templates, content calendar, posting cadence, and the exact frameworks they use - delivered to your inbox.
Posting patterns
Ng posts 3 times per week on average, with only 7 posts in the last 30 days, suggesting he prioritizes quality over frequency and may batch-create content around product launches.
Posts perform best on Sundays at 12 UTC (noon), which aligns with weekend reading habits when professionals have time to engage deeply.
Medium-length posts (80-180 words) average 5,056 engagements, outperforming very long posts (350+ words, 3,714 avg). However, his top posts are longer because they combine announcement + problem + solution + resources. The sweet spot is 150-250 words: long enough to establish credibility, short enough to hold attention.
Engagement by day
Content mix
avg 3.9K engagements per post
Ng's 69% problem-solution format works because his audience (software engineers) are pragmatists who want tools, not inspiration. By announcing a tool, explaining the problem it solves, and providing the installation command in one post, he collapses the gap between awareness and adoption. His 28% focus on AI agent tools and 23% on no-code education reflects his business (DeepLearning.AI, AI Fund) while his 18% on job market reality checks builds trust by addressing his audience's real anxieties rather than selling them hype.
