The AI Cookbook: AI Tools | Enterprise AI | Leadership

podcast artwork

Podcast by Malcolm Werchota

The AI Cookbook: AI Tools | Enterprise AI | Leadership

Malcolm Werchota's AI Cookbook is where artificial intelligence meets authentic business transformation. Known for his direct style and willingness to show AI in action—even during live presentations—Malcolm helps organizations understand that AI isn't about replacing humans but amplifying their capabilities. From voice-note productivity hacks to real-time meeting intelligence, this podcast delivers actionable insights for immediate implementation.

Latest episodes

episode artwork

19 October 2025

E85: Chip Shortage 2025: How Netherlands-China Dispute Stopped Cars

When the Dutch government seized chipmaker Nexperia on September 30, 2025, they thought they were protecting European interests. Instead, they triggered a supply chain catastrophe that could shut down automotive production worldwide.

In this episode of The AI Cookbook, Malcolm Werchota breaks down the spectacular policy failure that cut off 6 billion chips per month, threatens car factories across three continents, and reveals everything wrong with Europe's approach to the US-China tech war.

What You'll Learn:

How a single Dutch government action triggered China to block 6 billion chips per month - 40% of the global automotive semiconductor supply Why modern cars need 1,000+ chips and can't simply switch suppliers (qualification takes 6-18 months) The geopolitical chess match: US export controls, Dutch seizure, Chinese retaliation - and why Europe got caught in the crossfire What "just-in-time manufacturing" really means when supply chains become weapons Why this crisis is a preview of AI infrastructure vulnerabilities in Europe

Why This Matters:

This isn't just about cars or semiconductors. Malcolm explains how the Nexperia crisis reveals the fragility of global supply chains in an era of great power competition - and what it means for anyone building AI infrastructure in a world where technology is increasingly weaponized.

"You can't seize technological sovereignty. You have to build it systematically. Europe just learned this lesson the hard way."

Perfect For:

Supply chain professionals, business leaders managing geopolitical risk, AI teams building European infrastructure, anyone navigating the US-China tech war.

KEY TOPICS COVERED

───────────────────────────────────────────────────────

The Nexperia Crisis Timeline (00:45) September 30: Netherlands seizes Nexperia under emergency powers October 4: China blocks chip exports from Guangdong facility October 17: Alliance for Automotive Innovation warns of US plant shutdowns 6 billion chips per month suddenly trapped - 40% of automotive semiconductor market BMW, Toyota, Mercedes-Benz, Volkswagen, Ford, GM, Stellantis all affected

Understanding the Supply Chain Catastrophe (03:20) Why you can't just switch chip suppliers: 6-18 month qualification process Modern cars require 1,000+ semiconductor components Nexperia's role: Basic transistors and diodes (not advanced AI chips) The problem: Design in Netherlands, manufacturing in Germany, assembly in China 80% of finished products now trapped behind Chinese export blocks

The Geopolitical Chess Match (05:45) US-China tech war escalation and semiconductor weaponization Europe caught between two superpowers - no good options Netherlands' miscalculation: Legal control ≠ operational control Why China targeted automotive chips for maximum pain The death of "strategic autonomy" as a policy framework

Supply Chain Vulnerabilities Exposed (08:30) Just-in-time manufacturing's incompatibility with geopolitical risk Why automotive chips became the battlefield Geographic dependencies and single points of failure The cost of efficiency without resilience How supply chains become weapons in great power competition

What This Means for AI Infrastructure in Europe (11:15) If we can't secure basic chips, how do we build AI capabilities? Practical strategies for managing geopolitical risk in technology Why resilience now matters more than efficiency Lessons for anyone building on global supply chains The new reality: Security and redundancy over cost optimization

───────────────────────────────────────────────────────

EPISODE INSIGHTS

───────────────────────────────────────────────────────

Malcolm's Core Message: "The Death of Strategic Autonomy"

Europe attempted to navigate the US-China tech war by seizing Nexperia to protect supply chains. Instead, they demonstrated why you can't build technological sovereignty through government seizures. Real resilience requires systematic investment in full-stack capabilities - fabrication, packaging, assembly, testing - built BEFORE taking actions that trigger retaliation.

The Nexperia crisis is a case study in what NOT to do. The question is whether we'll learn the lessons before the next crisis hits.

Notable Quote:

"This crisis exposes the fragility of globalized supply chains when caught between Washington and Beijing. Modern cars require over 1,000 semiconductor components. Qualifying new suppliers takes months or years. Just-in-time manufacturing means no backup inventory. The math is brutal: no chips = no cars."

───────────────────────────────────────────────────────

RESOURCES & LINKS MENTIONED

───────────────────────────────────────────────────────

Alliance for Automotive Innovation warning statement Dutch Goods Availability Act documentation US Entity List and export control regulations Nexperia company information and facilities

───────────────────────────────────────────────────────

WHERE TO FIND MALCOLM WERCHOTA

───────────────────────────────────────────────────────

LinkedIn: https://www.linkedin.com/in/malcolmwerchota/

Website: https://www.werchota.ai/

YouTube: https://www.youtube.com/@werchota

X: https://x.com/malcolmwerchota

Facebook: https://www.facebook.com/people/AI-Cookbook-by-Malcolm-Werchota/61580362300250/?sk=reels_tab

Instagram: https://www.instagram.com/malcolmwerchotaai/

TikTok: https://www.tiktok.com/malcolmwerchota

───────────────────────────────────────────────────────

GET IN TOUCH

───────────────────────────────────────────────────────

Have questions about AI implementation or want to share your transformation story? Reach Malcolm at malcolm@werchota.ai

───────────────────────────────────────────────────────

EMAIL THE SHOW

───────────────────────────────────────────────────────

Write to us with episode requests, feedback, or ideas at social@werchota.ai

───────────────────────────────────────────────────────

READY TO LEVEL UP YOUR AI SKILLS?

───────────────────────────────────────────────────────

Explore AI Fit Academy – Malcolm's program helping professionals and teams apply AI tools effectively in work and business. Ship First, Study Later – working workflows by Week 2 or 100% refund. Learn more at https://www.werchota.ai/ai-fit-academy

───────────────────────────────────────────────────────

TAGS FOR SOCIAL & SEARCH

───────────────────────────────────────────────────────

#ChipShortage #AutomotiveIndustry #SupplyChain #Semiconductors #Geopolitics #USChinaTechWar #Manufacturing #Nexperia #Netherlands #China #SupplyChainManagement #GeopoliticalRisk #AutomotiveCrisis #TechPolicy #AIInfrastructure

00:00

26:09

episode artwork

19 October 2025

E84 - AI Drama | Brazil's Lesbian Dating App Disaster: AI Security Flaw

🎧 Listen now:

Spotify:

https://open.spotify.com/episode/249ZA6nHHoKmaiGYqY6Jum?si=91mGWjWJT-ur14At1KWpjA

Apple Podcast

https://podcasts.apple.com/at/podcast/brazils-lesbian-dating-app-disaster-ai-security-flaw/id1846704120?i=1000732455609

💔 Description

Marina thought she finally found safety.

A lesbian dating app in Brazil — built by queer women, for queer women.

Manual verification. No fake profiles. No men.

Then everything went wrong.

In September 2025, Sapphos launched as a sanctuary with government-ID checks.

Within 48 hours, 40,000 women downloaded it.

A week later, a catastrophic flaw exposed the most sensitive data of 17,000 users — IDs, photos, names, birthdays.

🔍 One researcher discovered he could view anyone’s profile just by changing a number in a URL.

That’s how fast “safety” can vanish when speed beats security.

🧠 What This Episode Covers

This episode of AI Drama investigates how AI-generated code, underqualified devs, and “vibe coding” collided with a vulnerable community.

It’s not a takedown of two activists — it’s a warning about asking for extreme trust without professional security.

🎓 You’ll Learn

  • How a single IDOR-style bug leaked government IDs and photos
  • Why AI-generated code often ships with hidden flaws
  • The unique threats LGBTQ+ apps face in high-violence regions
  • What happened after the founders deleted evidence of the breach
  • How to spot red flags before uploading your ID anywhere

⚠️ The Real Stakes

🇧🇷 Brazil remains one of the most dangerous countries for LGBTQ+ people.

Lesbian and bisexual women face three times higher rates of violence than straight women.

For many Sapphos users, being outed wasn’t embarrassing — it was life-threatening.

🧩 What Went Wrong

  • Identity checks increased trust — but concentrated risk
  • When one app collects IDs, selfies, and locations, a single bug exposes everything
  • AI sped up insecure coding — ~45 % of AI-generated code has vulnerabilities
  • No audits, no penetration tests, poor access control
  • Logs deleted → evidence erased
  • Communication failed: instead of transparency, users saw silence and denial

🚨 Red Flags Before Trusting an App

✅ Verified security audits (SOC 2 / ISO 27001)

✅ Transparent privacy policy + deletion options

✅ Minimal data collection — no unnecessary IDs

✅ Public security contact or bug-bounty page

✅ Experienced, visible founding team

❌ Avoid apps claiming “100 % secure” or “completely private”

🛡️ Safer Habits

🔑 Use unique emails + a password manager

🕵️ Prefer privacy-preserving verification methods

📍 Turn off precise location & strip photo metadata

🆔 After any breach: change credentials, rotate IDs if possible, monitor credit

💬 Notable Quotes

“Marina’s only ‘mistake’ was trusting people who promised protection.”
“The lesson isn’t don’t build — it’s don’t build insecure. Demand proof, not promises.”

📊 Select Facts

  • ~45 % of AI-generated code shows security flaws
  • LGBTQ+ users face more online harassment
  • Brazil records one LGBTQ+ person killed every ~48 hours

🎙️ AI Drama is a narrative-journalism podcast about the human cost when technology fails those who trust it most.

Hosted by Malcolm Werchota.

🔍 SEO Keywords

dating-app breach • LGBTQ privacy • Brazil • ID verification • AI code security • queer safety

00:00

08:47

episode artwork

18 October 2025

E83 - Weekly News Recap - Google's Cancer Breakthrough & ChatGPT Updates

Google DeepMind's AI just discovered a new cancer treatment that scientists validated in the lab - and it could transform how we treat 85% of lung cancers. This is AI's moonshot moment.

In Episode 83 of The AI Cookbook, Malcolm Werchota breaks down the 5 biggest AI stories from October's second week, proving we're witnessing the industrialization of artificial intelligence. From breakthrough medical discoveries to tools that finally make AI adoption measurable, this weekly recap delivers the business insights you need in under 15 minutes.

What You'll Learn:

  • How Google's Cell-to-Sentence 27B model discovered a drug that makes "cold tumors" visible to immune systems, achieving 50% increase in antigen presentation
  • OpenAI's controversial policy shift allowing adult content generation - and what Sam Altman's "we're not the moral police" means for AI competition
  • Why Google is investing $15 billion in India's largest data center outside the US (hint: 1.4 billion potential users)
  • Claude Haiku 4.5: Anthropic's new model runs 4-5x faster than Sonnet, achieves 73% on coding benchmarks, and costs one-third as much
  • Microsoft Copilot's game-changing benchmarking tool in Viva Insights - finally, a measurable AI adoption score you can track against peer companies

Why This Matters:

Malcolm draws a powerful parallel to Ford's assembly line innovation. The cancer discovery proves AI can do breakthrough scientific work. The new tools - faster AI agents, adoption metrics, and massive infrastructure investments - show we now have everything needed to deploy AI at industrial scale.

"We're getting to a point where the validation of AI isn't 'Can you build me a nice itinerary for my vacation in Rome?' But can you make a new discovery that will save patients' lives?"

Perfect For:

Business leaders tracking AI adoption, professionals implementing AI tools, teams measuring AI ROI, anyone who needs practical AI insights without the hype.

KEY TOPICS COVERED

───────────────────────────────────────────────────────

  1. Google DeepMind's Cancer Treatment Discovery (00:45)
    • Cell-to-Sentence 27B model identifies new immunotherapy pathway
    • Drug silmitacertib makes "cold tumors" visible to immune system
    • 50% increase in antigen presentation validated experimentally
    • Target: small cell lung cancers (85% of all lung cancer cases)
    • Proves AI can make genuine scientific breakthroughs
  2. OpenAI's Adult Content Policy Reversal (03:20)
    • Following Grok/XAI's lead by allowing adult content generation
    • Sam Altman: "We're not the moral police"
    • Analysis of competitive pressure from less restrictive AI models
    • Privacy and ethical implications for enterprise AI adoption
  3. Google's $15 Billion India Data Center Investment (05:45)
    • One of Google's largest data centers outside the United States
    • Gigabyte-scale computing infrastructure
    • Power requirements equivalent to 750,000 homes
    • Strategic positioning for India's 1.4 billion population
    • What this infrastructure buildout signals about AI's future scale
  4. Anthropic Releases Claude Haiku 4.5 (07:30)
    • Runs 4-5x faster than Claude Sonnet 4.5
    • Achieves 73% on SWE coding benchmark
    • Optimized for AI agent orchestration and multi-agent platforms
    • Available on Claude Code, API, Amazon Bedrock
    • One-third the cost of Sonnet with near-equivalent performance
    • Why speed and cost matter for business AI adoption
  5. Microsoft Copilot Benchmarking Tool - Malcolm's Favorite (09:15)
    • New feature in Viva Insights platform
    • Creates measurable AI adoption scores for organizations
    • Benchmark your Copilot usage versus peer companies
    • Track adoption by region, job role, and compare to top performers
    • Finally: accountability and quantitative metrics for AI transformation
    • Why measurement is the missing piece for enterprise AI adoption

───────────────────────────────────────────────────────

EPISODE INSIGHTS

───────────────────────────────────────────────────────

Malcolm's Overarching Theme: "Industrialization of AI"

Malcolm compares this moment to Henry Ford's assembly line innovation - not because the technology is new, but because we're finally scaling proven capabilities for universal deployment. The Google cancer discovery validates that AI can do genuinely important work. The supporting stories (Haiku's speed, Microsoft's metrics, Google's infrastructure) prove we now have the tools to industrialize AI across every business function.

Notable Quote:

"We're getting to a point where the validation of AI isn't anymore 'Can you build me a nice itinerary for my vacation in Rome?' But can you make a new discovery that will save patients' lives? That's what we're seeing this week."

───────────────────────────────────────────────────────

RESOURCES & LINKS MENTIONED

───────────────────────────────────────────────────────

  • Google DeepMind Cell-to-Sentence 27B model announcement
  • Anthropic Claude Haiku 4.5 release documentation
  • Microsoft Viva Insights Copilot Dashboard
  • OpenAI content policy updates

───────────────────────────────────────────────────────

WHERE TO FIND MALCOLM WERCHOTA

───────────────────────────────────────────────────────

LinkedIn: https://www.linkedin.com/in/malcolmwerchota/

Website: https://www.werchota.ai/

YouTube: https://www.youtube.com/@werchota

X: https://x.com/malcolmwerchota

Facebook: https://www.facebook.com/people/AI-Cookbook-by-Malcolm-Werchota/61580362300250/?sk=reels_tab

Instagram: https://www.instagram.com/malcolmwerchotaai/

TikTok: https://www.tiktok.com/malcolmwerchota

───────────────────────────────────────────────────────

GET IN TOUCH

───────────────────────────────────────────────────────

Have questions about AI implementation or want to share your transformation story? Reach Malcolm at malcolm@werchota.ai

───────────────────────────────────────────────────────

EMAIL THE SHOW

───────────────────────────────────────────────────────

Write to us with episode requests, feedback, or ideas at social@werchota.ai

───────────────────────────────────────────────────────

READY TO LEVEL UP YOUR AI SKILLS?

───────────────────────────────────────────────────────

Explore AI Fit Academy – Malcolm's program helping professionals and teams apply AI tools effectively in work and business. Ship First, Study Later – working workflows by Week 2 or 100% refund. Learn more at https://www.werchota.ai/ai-fit-academy

───────────────────────────────────────────────────────

TAGS FOR SOCIAL & SEARCH

───────────────────────────────────────────────────────

#AICancerTreatment #GoogleDeepMind #MicrosoftCopilot #ClaudeAI #AIAdoption #AIBusinessTools #AnthropicClaude #AIAgents #OpenAI #AIMetrics #BusinessAI #AITransformation #PodcastAI #AINews #TechNews

00:00

20:27

episode artwork

16 October 2025

E82: Dubai's $1.4T AI Push: Strategic Expansion Insights from GITEX

Just back from GITEX Global 2025 in Dubai, our Managing Partner Sumeyra Yildirim noticed something fundamental: the conversation has shifted from “what can AI do?” to “how do we scale it?”—and the UAE ecosystem is deliberately optimized for speed.

In this episode, Malcolm Werchota breaks down why werchota.ai is seriously exploring expansion to the UAE, sharing insights from Dubai’s largest tech conference and the strategic logic behind this decision.

What you’ll discover:

  • The infrastructure reality: $1.4 trillion US investment framework, 500,000 Nvidia chips annually, and the 5-gigawatt Stargate UAE AI campus—50× larger than typical data centers.
  • Why the UAE government is racing toward becoming the world’s first AI-native government by 2027, with AI CEOs in every ministry since 2017.
  • How werchota.ai’s proven DACH market model could accelerate in an ecosystem where regulatory frameworks enable rather than slow deployment.
  • The competitive advantage of dual-market operations: learn fast in one market, apply lessons with proper governance in another.
  • Real examples of deployment timeline differences—three months in Dubai versus eighteen months in Germany for identical AI solutions.

Episode Summary:

Malcolm Werchota explores werchota.ai’s strategic consideration of UAE expansion, sharing data-driven insights from GITEX Global 2025. The episode examines how the UAE’s AI-optimized ecosystem—backed by $1.4 trillion in infrastructure investment, deliberate government policy, and rapid deployment frameworks—creates a unique environment for AI transformation at scale.

Key Topics Covered

  • Market shift: from “what can AI do?” to “how do we scale it?”
  • $1.4T UAE investment framework and US partnerships
  • 500,000 Nvidia AI chips annually
  • Stargate UAE: 5GW AI campus (50× typical data center)
  • Abu Dhabi’s $3.5B AI-native government strategy (by 2027)
  • Mohamed bin Zayed University of AI & K–12 mandatory AI education
  • GITEX 2025 highlights: autonomous governance, quantum AI, digital twins
  • Deployment timeline: 3 months (Dubai) vs 18 months (Germany)
  • Dual-market acceleration model
  • MGX sovereign wealth fund ($100B+ AI investments)
  • G42, Technology Innovation Institute, and UAE global AI partnerships

Notable Statistics

MetricFigure
UAE–US Investment Framework$1.4T (10 years)
ADNOC AI Expansion$440B
Nvidia Chips Imported500,000/year
Stargate AI Campus5GW (50× average DC)
MGX Fund Target$100B+
MGX Data Center Acquisition$40B
UAE Tech Funding (H1 2025)$1B
Abu Dhabi Gov’t Digital Strategy$3.5B (2025–2027)
Global AI Startup Funding Share53%
New Jobs from Copilot UAE152,000
AI GDP Target13.6% by 2031
Fully AI-native Government Target2027

Resources & References

UAE AI Initiatives:

  • GITEX Global 2025
  • UAE National AI Strategy 2031
  • Abu Dhabi Digital Strategy 2025–2027
  • Stargate UAE AI Campus

Organizations Mentioned:

G42, MGX, Mohamed bin Zayed University of AI, Technology Innovation Institute, ADNOC, Du

Global Partners:

OpenAI, Microsoft, Nvidia, Oracle, Cisco, SoftBank, BlackRock, Cerebras Systems

Contact & Links

Malcolm Werchota

🔗 LinkedIn

🌐 Website

🎥 YouTube

🐦 X (Twitter)

📘 Facebook

📸 Instagram

🎵 TikTok

Email Contacts:

  • General: malcolm@werchota.ai
  • GCC Inquiries: Sumeyra Yildirim (Managing Partner, GCC)
  • Show Feedback: social@werchota.ai

AI Fit Academy:

Learn AI tools hands-on. Ship First, Study Later — working workflows by Week 2 or 100% refund.

👉 Learn more

00:00

23:22

episode artwork

14 October 2025

E81: Build Better AI Agents (Part 2): The Five Building Blocks of Context Engineering

After learning why AI agents fail in Part 1 (attention budget, context rot, orchestration limits), Malcolm Werchota now reveals how to build scalable, long-running AI systems using Anthropic’s framework for Context Engineering.

This episode goes beyond prompts — it’s about architecture.

Malcolm introduces the five building blocks of Context Engineering:

1️⃣ System Prompts – Define your agent’s identity, purpose, core capabilities, and quality standards.

2️⃣ Minimal Tool Sets – Stop giving 20 tools; focus on what’s essential.

3️⃣ Just-in-Time Retrieval – Only load information when it’s needed, not everything at once.

4️⃣ Long-Horizon Strategies – Extend runtime with compaction, note-taking, and delegation.

5️⃣ Examples & Patterns – Train with diverse examples, anti-patterns, and confidence scoring.

Using practical cases from Werchota.ai — like invoice automation and large-scale feedback analysis — Malcolm demonstrates how these techniques turn fragile “demo agents” into reliable production-grade systems.

Key topics: agent architecture, context optimization, compaction, token management, orchestration patterns, Anthropic Claude Code implementation, and how to scale AI workflows in production environments.

Perfect for professionals working with Claude, GPT-5, or Gemini — and anyone ready to move from prompt engineering to system thinking.

🗒️ SHOW NOTES

Episode 81, Part 2: Build Better AI Agents Through Context Engineering

Malcolm Werchota breaks down the five practical building blocks of Context Engineering, showing how to design scalable AI systems that actually think ahead — not just follow commands.

WHAT YOU’LL LEARN

  • The five key building blocks of Context Engineering
  • How to write effective system prompts that guide decision-making
  • Why fewer tools = better agents
  • How to implement Just-in-Time data retrieval
  • Extending AI lifespan through compaction and delegation
  • Using examples and anti-patterns to improve agent reasoning
  • Confidence scoring and note-taking for long-running tasks

KEY TAKEAWAYS

  • System Prompts: Define identity, purpose, and quality — short and structured (600–800 tokens).
  • Minimal Tool Sets: Reduce decision complexity; fewer, focused tools improve speed and reliability.
  • Just-in-Time Retrieval: Load only what’s needed in context; one file or task at a time.
  • Long-Horizon Strategies: Use compaction, external note-taking, and delegation to prevent context overload.
  • Examples & Patterns: Teach your agents from both successes and failures — diversity beats volume.

REAL-WORLD USE CASES

  • Invoice automation using Claude Code orchestration
  • Customer feedback summarization (10,000 → 5,000 words)
  • Parallel sub-agent processing (reading 10 invoices simultaneously)
  • Long-running report generation using compaction & note-taking

TOOLS & PLATFORMS

  • Claude Code (Anthropic)
  • Claude Sonnet 4.5 (1M-token context window)
  • Gemini 2.5 (1M-token context window)
  • ChatGPT-5 (200k-token context window)
  • Werchota.ai Cloud Dashboard (Episode Notes)

RESOURCES

  • Anthropic Research: Effective Context Engineering for AI Agents
  • Previous Episode: Build Better AI Agents – Part 1 (Context Engineering Basics)
  • Claude Code Documentation
  • Werchota.ai Blog: “Context Engineering in Real Workflows”

MALCOLM’S KEY INSIGHTS

“Don’t give your agent 20 tools — it will spend half its energy deciding which one to use.”
“The future of AI isn’t about bigger models. It’s about better architecture and context engineering.”
“System prompts are not messages — they’re thinking frameworks.”
“Context engineering turns fragile demos into production systems.”

🔗 WHERE TO FIND MALCOLM WERCHOTA

LinkedIn → linkedin.com/in/malcolmwerchota

Website → werchota.ai

YouTube → youtube.com/@werchota

X → x.com/malcolmwerchota

Facebook → AI Cookbook by Malcolm Werchota

Instagram → @malcolmwerchotaai

TikTok → @malcolmwerchota

📧 Get in touch:

Questions, feedback, or transformation stories → malcolm@werchota.ai

Episode ideas → social@werchota.ai

🎓 Upgrade your AI skills:

Join the AI Fit Academy — Malcolm’s hands-on program that helps professionals and teams ship real AI workflows by Week 2 — or your money back.

Learn more → werchota.ai/ai-fit-academy

00:00

31:46

episode artwork

14 October 2025

E80: Build Better AI Agents: Context Engineering Over Prompts (Pt. 1)

Your AI agents work—but they’re not smart. They follow instructions, yet fail on edge cases, forget context mid-task, and need constant supervision.

Malcolm Werchota reveals why your invoice automation, podcast metadata generation, and business workflows keep breaking down—and why it’s not the AI’s fault. The missing piece is context engineering, a concept most people have never heard of.

In this two-part series, Malcolm breaks down Anthropic’s groundbreaking research on how to build AI agents that actually think ahead. Learn why prompt engineering is no longer enough, how attention budget silently kills your automations, what context rot does to long-running tasks, and how the orchestrator pattern allows AI agents to spawn helper agents on demand.

You’ll hear how Malcolm’s team cut invoice processing time from 45 minutes to zero human intervention—and why feeding your AI more data can actually make it dumber. This episode isn’t about magic prompts. It’s about designing the entire environment your AI operates in.

Key topics: AI agent automation challenges, context window vs. attention budget, why mega-prompts fail, orchestrator pattern design, system prompt architecture, tool-calling strategies, and scalable AI workflows.

Perfect for professionals implementing Claude AI, automating business processes, or frustrated with unreliable AI agents. Malcolm’s “Ship First, Study Later” approach means real implementation—not theory.

Part 2 dives into advanced system prompts, minimal tool sets, and managing long-running tasks without context explosion.

WHAT YOU’LL LEARN

  • Why functional AI agents still fail at business automation
  • The difference between prompt vs. context engineering
  • How attention budget and context rot sabotage your workflows
  • The orchestrator pattern: when agents build their own helpers
  • Real-world cases: invoices, podcasts, and process automation
  • Why mega-prompts make AI dumber—and what to do instead
  • Anthropic’s context engineering framework
  • How to design information architecture for Claude and other LLMs

TOOLS & PLATFORMS

  • Claude Code (Anthropic)
  • Claude Sonnet 4.5 (1M token window)
  • Gemini 2.5 (1M token window)
  • ChatGPT (100–200k token window)
  • 10 Valley OS (TenVOS) – context engineering case study

RESOURCES

  • Anthropic Research: Effective Context Engineering for AI Agents
  • Previous Episode: Building Claude Code Agents
  • Previous Episode: 10 Valley OS – Context Engineering in Action

MALCOLM’S KEY INSIGHTS

“It’s like having an employee who follows orders perfectly—but never takes initiative or thinks ahead.”
“Context engineering manages everything the model uses: system instructions, tools, message history—not just the prompt.”
“The challenge now isn’t crafting perfect prompts. It’s curating the information within the model’s limited attention budget.”
“Don’t feed it a billion files. Use the smallest, clearest, highest-signal inputs possible.”

COMING IN PART 2

  • Advanced system prompt structure
  • Minimal tool sets for reliability
  • Handling long-running tasks without context explosion
  • Practical implementation blueprints

🔗 WHERE TO FIND MALCOLM WERCHOTA

LinkedIn → linkedin.com/in/malcolmwerchota

Website → werchota.ai

YouTube → youtube.com/@werchota

X → x.com/malcolmwerchota

Facebook → AI Cookbook by Malcolm Werchota

Instagram → @malcolmwerchotaai

TikTok → @malcolmwerchota

📧 Get in touch:

Questions, feedback, or transformation stories → malcolm@werchota.ai

Episode ideas → social@werchota.ai

🎓 Upgrade your AI skills:

Check out the AI Fit Academy, Malcolm’s hands-on program that gets professionals shipping working AI workflows by Week 2—or your money back.

Learn more → werchota.ai/ai-fit-academy

00:00

26:35

Copyright © The AI Cookbook: AI Tools | Enterprise AI | Leadership. All rights reserved.

Powered by