Matt Pocock's LinkedIn Strategy, Author of…

Ways I'm using AI for coding: - Building Features: plan mode, multi-phase plans, review each phase - Fixing Bugs: background agent creates repro, background agent 2 fixes repro, review - Tweaking UI:…

28829128 viral

AI10 months ago

View on LinkedIn

Introducing LLM Fundamentals! AI is powerful, but weird. And understanding how works under the hood is one of the most powerful things you can do for your career. So, here's a 5-lesson FREE tutoria…

1509135 viral

AI11 months ago

View on LinkedIn

Explicitly spawning subagents with Claude Code is extremely freaking cool My workflow is: - Enter plan mode - Explicitly say how many subagents I want and which tasks they should perform - Let it ri…

LinkedIn post image: Explicitly spawning subagents with Claude Code is extremely freaking cool

157653 viral

AI11 months ago

View on LinkedIn

Tell your LLM to SACRIFICE GRAMMAR

150553 viral

AI11 months ago

View on LinkedIn

In 2025, the AI SDK went from 400k to 3.5 MILLION weekly downloads. And I've built a self-paced, interactive, 57-lesson course to help you master it. And for a few days, it's $99. 2025 is the year…

119153 viral

AI Development10 months ago

View on LinkedIn

The lesson that never stops being true with AI: ASSUME THE AI KNOWS NOTHING. Work backwards from your desired output to find exactly what pieces of context the AI needs.

LinkedIn post image: The lesson that never stops being true with AI:

110353 viral

AI11 months ago

View on LinkedIn

Topics & Content Focus

Primary Topics

Practical AI-assisted software development workflowsTypeScript-centric AI tooling and the AI SDK ecosystemAgentic patterns for LLMs (retrieval, memory, evals, human-in-the-loop)Developer-focused courses and cohort-based AI education

Secondary Themes

Developer productivity and code automation with AIHands-on experimentation with AI coding tools (Claude Code and similar)Offer design and pricing for technical education productsBridging traditional TypeScript skills into AI application development

Industry Focus

Developer tooling and AI infrastructureAI/ML developer education and upskillingTechnical online education and cohort-based learning products

Content Categories

Educational how-to breakdowns of AI coding workflowsProduct and course launch announcementsTool-specific workflow demos and mini case studiesSales and promotional posts tied to discounts and limited-time offers

Performance Insights

10%

Avg Engagement Rate

STABLE

Performance Trend

Best Performing Topics

Concrete, step-by-step AI coding workflows and personal usage patternsDeep dives into specific AI coding assistants and agent configurationsEducational content that connects AI SDK concepts to real build outcomes

Virality Signals

Posts that expose specific personal workflows ('Ways I'm using AI for coding', 'My workflow is') generate disproportionately high comments and sharesEducational posts that show how to combine agents, retrieval, memory, and evals perform better than pure promotionsMetric-backed credibility (download counts, 'deepest discount ever') boosts interest but needs practical detail to spike engagementTool-specific posts that reveal non-obvious uses (e.g. explicitly spawning subagents) are highly shareable within developer networks

Structure & Quality

Avg Length (Words)

MEDIUM

Depth Level

ADVANCED

Expertise Level

8/10

Uniqueness Score

Common Hooks

Outcome-first hooks that immediately frame who the offer is for (e.g. 'Folks who have been waiting to buy…')Time- or cohort-based announcement hooks (e.g. 'NEW COHORT 💥')Metric-driven credibility hooks (e.g. '[tool] went from 400k to 3.5 MILLION weekly downloads')Personal workflow reveals (e.g. 'Ways I'm using AI for coding:', 'My workflow is:')

Common Endings

Implied urgency around timing or pricing (e.g. 'for a few days, it's $99', 'now's the moment')Short motivational push (e.g. 'Let's go.')Clear statement of what the reader will build or learn (e.g. 'You'll build:' lists)

Value Delivery Methods

Shares concrete, reproducible AI workflows that developers can immediately tryClarifies complex AI concepts (agents, evals, retrieval, memory) through applied examplesConnects tools and techniques directly to what learners will build, not just abstract theoryProvides clear pathways from free/entry products to deeper, structured learning (crash course → cohort)

Formatting Style

Short paragraphs separated by blank lines for scanabilityHeavy use of bullet lists for 'You'll learn / You'll build' or step-by-step workflowsOccasional all-caps for emphasis (e.g. 'EVERYTHING', 'NEW COHORT')Section headers with labels (e.g. 'DAY 1:', 'DAY 2:')Sparse emojis used mainly in announcement hooks (e.g. 💥)

Audience & Tone

YES

Question Usage

0.5%

Response Rate

Detected Tone

Relatable expert builderPragmatic and implementation-focusedSmart-casual with occasional hype momentsTechnical but accessible to practicing developerssemi-formalfirst-person

Interaction Style

Primarily broadcast-style posts with strong informational valueImplicitly conversational through 'you' framing and direct address to developersEngagement driven more by resonance with workflows and tools than by explicit prompts for discussion

Community Building Signals

Rewards existing learners with financial incentives to move up the product ladder (e.g. course credit toward cohorts)Positions followers as part of a broader movement ('2025 is the year of TypeScript + AI')Builds a recurring narrative around mastering a specific ecosystem (AI SDK + TypeScript) over time

Writing Style Patterns

Content Strategy

Hook: Outcome-first hooks that immediately frame who the offer is Tone: semi-formalCTA: Direct purchase CTAs coupled with scarcity/discoun

Writing style breakdown

Core characteristics

Highly concise, punchy, and direct.

Professional but casual; confident and a bit playful.

Feels like a mix of product marketing + dev educator + Twitter-style one-liners.

- Zero fluff. Almost every sentence either

Announces something.

Teaches a distilled principle.

Describes a workflow or feature.

Acts as a hook.

Formality

Semi-informal, but not slang-heavy.

Uses contractions naturally ("We'll", "It's", "don't"), but avoids chatty filler.

Feels like a technically competent founder/engineer talking to peers, not like a corporate press release.

Energy and tone

High energy, but controlled.

Uses exclamation marks sparingly but strategically (e.g. "NEW COHORT 💥", "Introducing LLM Fundamentals!").

Tone is optimistic, forward-looking, and slightly hype-driven, but still grounded in specifics and utility.

Very confident; never tentative or hedged.

Signature traits

Strong hooks in 1–2 lines at the very start.

- Frequent use of all-caps for emphasis on a key phrase or core principle

EVERYTHING is 50% off

NEW COHORT

ASSUME THE AI KNOWS NOTHING.

SACRIFICE GRAMMAR

- One-line maxims or slogans

None of this is vibe coding

Make Claude Code give you answers, not essays

The hidden security risk in your system prompts

Very compact lists of topics or steps, especially when teaching.

- Occasional light humor or playfulness

MOAR Retrieval

stupid model, fast loop

Extremely freaking cool

- Reader is addressed mostly in second person ("you") when

Selling something ("Ever wanted to customize an LLM? This one's for you.")

Advising ("Tell your LLM to SACRIFICE GRAMMAR")

Teaching ("understanding how works under the hood is...for your career.")

- First person ("I", "we") is used

To describe what the author is doing/building ("We're building...", "I've built...", "Ways I'm using AI for coding:")

To create social proof or authority.

- Commands are usually direct and imperative

Tell your LLM to SACRIFICE GRAMMAR

Make Claude Code give you answers, not essays

Suggestions are rarely soft. The voice assumes the reader wants clear direction.

Matt Pocock

Warm Analysis

Performance Overview

Top Posts by Engagement

Posting Patterns & Frequency

Best Performing Days

Best Performing Times To Post