AI & ChatGPT

Matt Pocock (AI hero) – Build Deep Search in TypeScript

This is a cohort-based course and the lessons will start unlocking on July 14th, 2025.

Intermediate English 4h 10m Matt Pocock (AI hero)
0:00 / 0:00

What you’ll learn

  • Implement essential backend infrastructure (databases, caching, auth) specifically for AI-driven applications.
  • Debug and understand the “black box” of AI agent decisions, especially when multiple tools are involved.
  • Ensure chat persistence, reliable routing, and real-time UI updates for a seamless user experience.
  • Objectively measure AI performance moving beyond subjective “vibe checks” for improvements.
  • Manage complex agent logic without creating brittle, monolithic prompts that are hard to maintain and optimize.
  • Understand the key differences between proof-of-concept AI apps and production-ready AI applications.

Overview

This is a cohort-based course and the lessons will start unlocking on July 14th, 2025.

Building AI applications that are genuinely useful involves more than just hitting an LLM API and getting back stock chat responses.

The difference between a proof-of-concept and a production application lies in the details.

Generic chat responses might work for demos, but professional applications need appropriate outputs that align with specific requirements.

In a professional environment code is (ideally) tested, metrics are collected, analytics are displayed somewhere.

AI development can follow these established patterns.

You will hit roadblocks when trying to:

  • Implement essential backend infrastructure (databases, caching, auth) specifically for AI-driven applications.
  • Debug and understand the “black box” of AI agent decisions, especially when multiple tools are involved.
  • Ensure chat persistence, reliable routing, and real-time UI updates for a seamless user experience.
  • Objectively measure AI performance moving beyond subjective “vibe checks” for improvements.
  • Manage complex agent logic without creating brittle, monolithic prompts that are hard to maintain and optimize.

In this course you will build out a “DeepSearch” AI application from the ground up to help you understand and implement these patterns and ensure a production-ready product.

What Will I Learn?

  • Understand the key differences between proof-of-concept AI apps and production-ready AI applications.

  • Implement backend infrastructure tailored for AI-driven apps, including databases, caching, and authentication.

  • Debug and interpret AI agent decision-making, even when using multiple tools.

  • Build reliable chat persistence, routing, and real-time UI updates for a smooth user experience.

  • Measure AI performance with objective metrics instead of relying on subjective impressions.

  • Manage complex AI agent logic without creating brittle, hard-to-maintain prompts.

  • Develop a fully functional “DeepSearch” AI application from scratch, following professional development patterns.

Target Audience

  • Developers who want to move from AI prototypes to scalable, production-ready applications.

  • Engineers and tech leads looking to integrate AI into real-world products with reliability and performance tracking.

  • Backend and frontend developers who want to understand AI-specific infrastructure needs.

  • Product managers or technical founders aiming to create polished AI tools beyond generic chatbots.

Requirements / Instructions

  • Basic programming skills (JavaScript, Python, or similar).

  • Understanding of web application development fundamentals.

  • Willingness to work with both frontend and backend technologies.

  • Interest in debugging and improving AI systems beyond surface-level outputs.

  • Commitment to follow the course schedule as lessons unlock starting July 14th, 2025.

Materials included

  • 113 on-demand HD video lessons (4h 10m)
  • 2 structured curriculum modules
  • Full offline course archive download included
  • Instant classroom streaming on desktop, tablet, and mobile

Requirements

  • No prior advanced technical setup required — lessons progress step by step
  • A desktop, laptop, or mobile browser for HD video streaming and workbook access

Target audience

  • Entrepreneurs and consultants automating workflows and launching AI services
  • Creators and marketers using ChatGPT, custom GPTs, and AI agents for leverage
  • Non-technical builders shipping AI-powered products and client deliverables

Curriculum

Matt Pocock AIhero - Build DeepSearch in TypeScript

Module containing 113 video lessons

  • Lesson 1: What Are We Building Preview 02:10
  • Lesson 2: Installation Instructions Dont Skip This 03:19
  • Lesson 3: Cursor Tips 00:23
  • Lesson 4: Explore The Repo Problem 02:00
  • Lesson 5: Explore The Repo Solution 02:59
  • Lesson 6: Setting Up Postgres 02:01
  • Lesson 7: Using Drizzle And Drizzle Studio 02:00
  • Lesson 8: Setting Up Redis 02:00
  • Lesson 9: Faqs 02:00
  • Lesson 10: Introduction 02:00
  • Lesson 11: Choose An Llm Problem 02:00
  • Lesson 12: Choose An Llm Solution 02:00
  • Lesson 13: Our First Model Call Problem 02:00
  • Lesson 14: Our First Model Call Solution 02:00
  • Lesson 15: Set Up Discord Authentication Problem 02:00
  • Lesson 16: Set Up Discord Authentication Solution 02:00
  • Lesson 17: Create A Naive Agent With Serper Problem 02:12
  • Lesson 18: Create A Naive Agent With Serper Solution 02:00
  • Lesson 19: Showing Tool Calls In The Frontend Problem 02:29
  • Lesson 20: Showing Tool Calls In The Frontend Solution 02:00
  • Lesson 21: Search Grounding Optional Problem 02:00
  • Lesson 22: Search Grounding Optional Solution 02:01
  • Lesson 23: Rate Limiting Optional Problem 02:00
  • Lesson 24: Rate Limiting Optional Solution 02:00
  • Lesson 25: Connecting Our App To Mcp Servers Optional 02:00
  • Lesson 26: Create Database Resources For Persisting Messages Problem 02:03
  • Lesson 27: Create Database Resources For Persisting Messages Solution 02:00
  • Lesson 28: Persist Chats To The Database Problem 02:20
  • Lesson 29: Persist Chats To The Database Solution 02:00
  • Lesson 30: Creating New Chats In The Frontend Problem 02:00
  • Lesson 31: Creating New Chats In The Frontend Solution 02:00
  • Lesson 32: Showing The Saved Chats In The Frontend Problem 02:00
  • Lesson 33: Showing The Saved Chats In The Frontend Solution 02:00
  • Lesson 34: Fixing The New Chat Button Optional Problem 02:00
  • Lesson 35: Fixing The New Chat Button Optional Solution 02:00
  • Lesson 36: Adding Use Scroll To Bottom Optional Problem 02:00
  • Lesson 37: Adding Use Scroll To Bottom Optional Solution 02:00
  • Lesson 38: Choosing An Observability Platform 02:26
  • Lesson 39: Integrating Langfuse Problem 02:00
  • Lesson 40: Integrating Langfuse Solution 02:00
  • Lesson 41: Passing Extra Metadata To Langfuse Problem 02:00
  • Lesson 42: Passing Extra Metadata To Langfuse Solution 02:00
  • Lesson 43: Adding A Scraper Problem 02:23
  • Lesson 44: Adding A Scraper Solution 02:06
  • Lesson 45: Making The Llm Date Aware Optional Problem 02:00
  • Lesson 46: Making The Llm Date Aware Optional Solution 02:00
  • Lesson 47: Improving Our Crawler Optional 02:44
  • Lesson 48: Reporting Db Calls To Langfuse Optional Problem 02:00
  • Lesson 49: Reporting Db Calls To Langfuse Optional Solution 02:00
  • Lesson 50: Initializing Evalite Problem 02:04
  • Lesson 51: Initializing Evalite Solution 02:00
  • Lesson 52: Choosing Our Success Criteria 04:09
  • Lesson 53: Making Our System Testable Problem 02:00
  • Lesson 54: Making Our System Testable Solution 02:00
  • Lesson 55: Our First Deterministic Eval Problem 02:00
  • Lesson 56: Our First Deterministic Eval Solution 02:00
  • Lesson 57: Adding A Global Rate Limiter Optional Problem 02:00
  • Lesson 58: Adding A Global Rate Limiter Optional Solution 02:09
  • Lesson 59: The Data Flywheel 02:42
  • Lesson 60: Our First Llm As A Judge Eval Problem 02:59
  • Lesson 61: Our First Llm As A Judge Eval Solution 02:00
  • Lesson 62: Create A Simple Dataset Problem 03:20
  • Lesson 63: Create A Simple Dataset Solution 04:08
  • Lesson 64: Organizing Our Dataset Into Dev Ci And Regression Optional Problem 02:00
  • Lesson 65: Organizing Our Dataset Into Dev Ci And Regression Optional Solution 02:00
  • Lesson 66: Assessing Answer Relevancy Optional Problem 02:00
  • Lesson 67: Assessing Answer Relevancy Optional Solution 02:00
  • Lesson 68: Extracting The Parameters Of Our System Optional Problem 02:00
  • Lesson 69: Extracting The Parameters Of Our System Optional Solution 02:00
  • Lesson 70: What S Wrong With Our Current Approach 02:00
  • Lesson 71: Designing Our New System Prompt Problem 02:00
  • Lesson 72: 1 Designing Our New System Prompt Problem 02:00
  • Lesson 73: Designing Our New System Prompt Solution 02:00
  • Lesson 74: Creating A Next Action Picker Problem 02:00
  • Lesson 75: Creating A Next Action Picker Solution 02:00
  • Lesson 76: Implementing The Loop Solution 02:00
  • Lesson 77: Connecting Our Loop To The Frontend Problem 02:00
  • Lesson 78: Connecting Our Loop To The Frontend Solution 02:00
  • Lesson 79: Smoothing Our Streaming Optional Problem 02:00
  • Lesson 80: Smoothing Our Streaming Optional Solution 02:00
  • Lesson 81: Showing The Steps Taken In The Frontend Problem 02:30
  • Lesson 82: Showing The Steps Taken In The Frontend Solution 02:00
  • Lesson 83: Fixing Telemetry Problem 02:00
  • Lesson 84: Fixing Telemetry Solution 02:00
  • Lesson 85: Passing The Message History Problem 02:00
  • Lesson 86: Passing The Message History Solution 02:00
  • Lesson 87: Persisting Our New Setup To The Backend Problem 02:00
  • Lesson 88: Persisting Our New Setup To The Backend Solution 02:38
  • Lesson 89: Generating Chat Titles Optional Problem 02:00
  • Lesson 90: Generating Chat Titles Optional Solution 02:00
  • Lesson 91: Adding Geolocation Info To The System Prompt Optional Problem 02:00
  • Lesson 92: Adding Geolocation Info To The System Prompt Optional Solution 02:00
  • Lesson 93: Agents Vs Workflows 02:38
  • Lesson 94: Collapse Search And Crawl Into One Tool Problem 02:00
  • Lesson 95: Collapse Search And Crawl Into One Tool Solution 02:00
  • Lesson 96: Search Scrape Summarize Problem 02:00
  • Lesson 97: Search Scrape Summarize Solution 03:19
  • Lesson 98: Making A Query Rewriter Problem 03:10
  • Lesson 99: Making A Query Rewriter Solution 02:38
  • Lesson 100: Use A Combined Search Scrape Api Instead Optional Problem 02:00
  • Lesson 101: Use A Combined Search Scrape Api Instead Optional Solution 02:00
  • Lesson 102: Resumable Streams Optional 03:59
  • Lesson 103: Building An Evaluator Problem 02:15
  • Lesson 104: Building An Evaluator Solution 02:05
  • Lesson 105: Showing Sources In The Frontend Problem 02:00
  • Lesson 106: Showing Sources In The Frontend Solution 02:00
  • Lesson 107: Implementing Guardrails Optional Problem 02:14
  • Lesson 108: Implementing Guardrails Optional Solution 02:00
  • Lesson 109: Implement An Ask Clarifying Questions Step Optional Problem 02:00
  • Lesson 110: Implement An Ask Clarifying Questions Step Optional Solution 02:00
  • Lesson 111: Showing Usage In The Frontend Optional Problem 02:00
  • Lesson 112: Showing Usage In The Frontend Optional Solution 02:00
  • Lesson 113: Migrating To Ai Sdk V 5 Optional 07:40
Download Full Course Resources
  • Full Course Archive & Offline Download
Full Course Archive & Offline Download

Download the complete course archive and reference files.

  • Full Course Archive & Offline Download

Course FAQ & key details

What will I learn in Matt Pocock (AI hero) – Build Deep Search in TypeScript?

This is a cohort-based course and the lessons will start unlocking on July 14th, 2025. Key topics include: Implement essential backend infrastructure (databases, caching, auth) specifically for AI-driven applications.; Debug and understand the “black box” of AI agent decisions, especially when multiple tools are involved.; Ensure chat persistence, reliable routing, and real-time UI updates for a seamless user experience..

How many modules and lessons are in Matt Pocock (AI hero) – Build Deep Search in TypeScript?

Matt Pocock (AI hero) – Build Deep Search in TypeScript is organized into 3 structured modules and 115 on-demand lessons (4h 10m) in the AI & ChatGPT category.

How do I access Matt Pocock (AI hero) – Build Deep Search in TypeScript after checkout?

Immediately after payment verification on IMCourse24h, you can stream lessons 24/7 in the online classroom with saved progress and access any included downloadable archives from Account → Downloads.

Is Matt Pocock (AI hero) – Build Deep Search in TypeScript available individually or through membership?

You can pay $50.00 USD for Matt Pocock (AI hero) – Build Deep Search in TypeScript as a single course, or buy an IMCourse24h All-Access Membership ($100 for 1 month, $200 for 1 year, or $300 one-time for lifetime) for complete A-to-Z access to all 600+ courses.

Customer reviews

No approved reviews yet.

Related courses

More practical training in AI & ChatGPT.

All AI & ChatGPT courses