## 📖 Quick Navigation - [What is this guide about?](#what-is-this-guide) - [TL;DR - 30 Second Summary](#tldr) - [Full Table of Contents](#full-table-of-contents) - [Start Learning](#ai-as-colleague) --- ## What is this guide about? **This is the guide between understanding AI and actually using it at work.** You now know what Generative AI is, how LLMs work, what they can and can't do, how to think with the 4D framework, and how to write effective prompts. What you may not have yet is a clear, honest picture of how all of that lands in your actual job — your meetings, your emails, your reports, your decisions. **What AI tasks save real time — and which ones create more work than they save?** That is what this guide answers. Real use cases, organised by task type and role, with clear guidance on where AI genuinely helps and where it needs your supervision. **In one sentence:** This guide gives you a working map of how AI fits into your daily work — so you stop wondering if you should use it and start knowing exactly when and how. --- ## TL;DR - 30 Second Summary AI in the workplace works best when it acts as a force multiplier on work you already understand: 1. **Draft** — let AI write the first version, you refine it 2. **Summarise** — let AI compress long content, you verify the key points 3. **Research** — let AI surface options, you apply judgment 4. **Explain** — let AI translate complexity, you validate accuracy 5. **Automate** — let AI handle repetitive structure, you handle decisions ```mermaid flowchart LR A["🧑 You\nExpertise + Judgment"] --> B["🤝 AI\nSpeed + Scale"] B --> C["✅ Output\nFaster · Clearer · More consistent"] D["⚠️ Without judgment"] --> E["❌ Unverified AI output\nConfident but wrong"] F["⚠️ Without AI"] --> G["🐌 Manual work\nSlow · Inconsistent"] style A stroke:#2980b9 style B stroke:#27ae60 style C stroke:#4a90e2 style D stroke:#e67e22 style E stroke:#c0392b style F stroke:#e67e22 style G stroke:#c0392b ``` **Key idea:** AI at work is not about replacing what you do. It is about removing the friction between what you know and what you produce — so your expertise reaches further, faster. --- ## Full Table of Contents 1. [AI as a Colleague — The Right Mental Model](#ai-as-colleague) 2. [The 5 Work Modes — How AI Helps at Work](#work-modes) 3. [Use Cases by Task — The Practical Map](#task-map) 4. [Use Cases by Role — Where You Fit In](#role-map) 5. [The Human-AI Handoff — Who Does What](#handoff) 6. [Common Mistakes — What Goes Wrong at Work](#mistakes) 7. [Your AI Workflow — Building a Personal System](#workflow) [Final Mental Model](#final-mental-model) | [Concept Reference Card](#concept-reference-card) ---

# AI in the Workplace ### Practical First. Role-Specific Second. Then Your Personal System. --- # 🌟 The Only Idea You Need AI at work is not a replacement. It is not magic. It is: 👉 **A very fast, very capable junior colleague who never gets tired — but always needs your review.** Give it clear instructions. Check its work. Take responsibility for the output. That is the entire model. --- # 🧑‍💼 1. AI as a Colleague — The Right Mental Model ## 🧠 Core Intuition **Problem:** Most people approach AI at work in one of two wrong ways — either they trust it completely and stop checking, or they distrust it completely and never use it. **What happens:** Both extremes cost you. The right mental model sits in the middle. Think of AI as a new colleague who: - Is extremely fast and never complains about volume - Has read almost everything ever written - Has no memory of what you talked about yesterday - Sounds confident even when they're wrong - Needs clear briefs to do good work - Should never sign off on something without your review ```mermaid flowchart TD A["🧑 You give a clear brief"] A --> B["🤖 AI produces a first draft"] B --> C{"✅ Review:\nAccurate? Complete? On-brand?"} C -->|"Yes"| D["📤 Use it"] C -->|"Needs work"| E["✏️ Edit and refine"] C -->|"Wrong"| F["🔁 Re-prompt or redo"] style A stroke:#2980b9 style B stroke:#27ae60 style C stroke:#f39c12 style D stroke:#4a90e2 style E stroke:#e67e22 style F stroke:#c0392b ``` **Why it matters:** The colleague mental model sets the right defaults — use AI actively, but never abdicate judgment. Every output that leaves your desk has your name on it, not AI's. **One-line memory:** 👉 AI at work = **a fast, capable colleague whose work always needs your sign-off** ---

⚙️ Technical Deep Dive (Optional)

**What is it called?** 👉 **Human-in-the-loop** — the principle that AI systems in consequential contexts should always have a human reviewing outputs before they are acted on **Why the colleague model works:** - It sets correct trust calibration — not zero, not absolute - It frames your role as editor and decision-maker, not typist - It makes the division of labour clear: AI handles volume and speed, you handle accuracy and judgment - It scales — the same mental model works whether you're writing an email or building an AI-powered product **How dots connect:** - Hallucination (from How LLMs Work) is why review is non-negotiable - Discernment and Diligence (from How to Think With AI) are the human competencies that make this model work - Prompt engineering (from Effective Prompt Engineering) is how you write the clear brief that gets good output

--- # ⚙️ 2. The 5 Work Modes — How AI Helps at Work ## 🧠 Core Intuition **Problem:** "Use AI at work" is too vague to act on. You need to know *what kind* of work AI helps with — and how. **What happens:** Almost every AI workplace use case falls into one of five modes. ```mermaid flowchart LR A["🤖 AI Work Modes"] --> B["📝 Draft\nFirst version of anything written"] A --> C["📋 Summarise\nCompress long content to key points"] A --> D["🔍 Research\nSurface options, ideas, background"] A --> E["💬 Explain\nTranslate complexity into clarity"] A --> F["🔁 Transform\nConvert content from one form to another"] style A stroke:#2980b9 style B stroke:#27ae60 style C stroke:#f39c12 style D stroke:#8e44ad style E stroke:#e67e22 style F stroke:#c0392b ``` **The five modes:** | Mode | What AI does | You do | |---|---|---| | **Draft** | Writes the first version | Edit, refine, verify, approve | | **Summarise** | Compresses content to essentials | Check nothing important was lost | | **Research** | Surfaces relevant ideas, options, examples | Apply judgment, verify facts | | **Explain** | Translates complex content into plain language | Validate accuracy, adjust for audience | | **Transform** | Converts content from one format to another | Review the output format | **Why it matters:** Naming the mode before you prompt improves your results immediately. Instead of "help me with this report," you know to say "draft an executive summary" (Draft mode) or "turn these bullet points into a narrative" (Transform mode). **One-line memory:** 👉 Five modes: **Draft · Summarise · Research · Explain · Transform** ---

⚙️ Technical Deep Dive (Optional)

**What is it called?** 👉 **Task decomposition** — breaking work into clearly typed subtasks before prompting **Why modes matter technically:** - LLMs respond differently depending on how the task is framed - A "draft" prompt activates generative behaviour — open-ended output - A "summarise" prompt activates compression behaviour — focused extraction - Mismatched mode and prompt = lower quality output every time **Combining modes:** Many real work tasks chain modes together: ``` Research → Draft → Transform "What are the key arguments for X?" (Research) → "Write a proposal section covering those arguments" (Draft) → "Now convert this into 5 bullet points for a slide" (Transform) ``` **How dots connect:** - Chain-of-thought prompting (from Effective Prompt Engineering) maps directly onto chaining modes - The 4D framework's Delegation competency (from How to Think With AI) is about recognising which mode to hand off to AI - Context window limits (from What is Generative AI?) affect how much you can do in a single prompt chain

--- # 🗺️ 3. Use Cases by Task — The Practical Map ## 🧠 Core Intuition **Problem:** Knowing the five modes still leaves a gap — what actual work tasks do those modes apply to? **What happens:** Here is the honest, practical map of AI task fit across common workplace activities. ```mermaid flowchart TD TASKS["📋 Workplace Tasks"] --> HIGH["✅ High Fit\nUse confidently"] TASKS --> MED["⚠️ Medium Fit\nUse carefully + verify"] TASKS --> LOW["❌ Low Fit\nDon't rely on AI"] HIGH --> H1["📧 Drafting emails & messages"] HIGH --> H2["📄 Summarising documents & meetings"] HIGH --> H3["📊 Writing reports & proposals"] HIGH --> H4["💡 Brainstorming ideas & options"] HIGH --> H5["🔄 Reformatting & restructuring content"] HIGH --> H6["📖 Explaining concepts in plain language"] MED --> M1["🔍 Researching background context"] MED --> M2["📐 Analysing qualitative feedback"] MED --> M3["🧾 Drafting contracts or policies"] MED --> M4["📊 Interpreting data or reports"] LOW --> L1["⚖️ Final legal or compliance decisions"] LOW --> L2["🔢 Live data lookups or real-time facts"] LOW --> L3["🧮 Complex numerical calculations"] LOW --> L4["🤝 Human judgment calls — hiring, firing, escalation"] style HIGH stroke:#27ae60 style MED stroke:#f39c12 style LOW stroke:#c0392b ``` **Why it matters:** Matching the task to the fit level before you start saves time and prevents costly mistakes. High-fit tasks: hand off freely and edit. Medium-fit tasks: use AI as a starting point, verify thoroughly. Low-fit tasks: use AI to support your thinking, not replace your decision. **One-line memory:** 👉 High fit = draft freely · Medium fit = verify thoroughly · Low fit = AI assists, you decide ---

⚙️ Technical Deep Dive (Optional)

**What determines task fit?** | Factor | High fit | Low fit | |---|---|---| | **Requires live data?** | No — evergreen content | Yes — real-time information | | **Requires precise facts?** | No — structure and language | Yes — specific numbers, dates, names | | **Consequential if wrong?** | Low — internal draft | High — legal, financial, safety | | **Verifiable by you?** | Yes — you can check it | No — you don't have the knowledge to verify | | **Structured pattern?** | Yes — emails, reports, summaries | No — novel judgment calls | **The verification principle:** The higher the consequence of an error + the harder it is for you to verify + the more specific the facts required = the lower the fit. **How dots connect:** - Task-to-AI fit (from AI Capabilities, Limitations & Use Cases) is the upstream framework this section applies - Hallucination risk is highest in Medium and Low fit tasks — this is where your Diligence competency matters most

--- # 👥 4. Use Cases by Role — Where You Fit In ## 🧠 Core Intuition **Problem:** Generic AI use cases don't always feel relevant to your specific job. **What happens:** Here is how the same AI capabilities land differently depending on your role. ```mermaid flowchart LR AI["🤖 AI Capabilities"] --> MGR["👔 Manager\nBriefs · Feedback · Decisions"] AI --> ANA["📊 Analyst\nSummarise · Structure · Explain data"] AI --> MKT["📣 Marketer\nCopy · Campaigns · Content"] AI --> ENG["💻 Engineer / Developer\nCode · Debug · Docs"] AI --> OPS["⚙️ Operations\nProcesses · SOPs · Templates"] AI --> HR["🧑‍🤝‍🧑 HR / L&D\nJob descriptions · Training materials"] style AI stroke:#2980b9 style MGR stroke:#8e44ad style ANA stroke:#f39c12 style MKT stroke:#27ae60 style ENG stroke:#4a90e2 style OPS stroke:#e67e22 style HR stroke:#c0392b ``` **Role-specific use cases:** | Role | Top 3 AI Use Cases | |---|---| | **Manager** | Draft performance feedback · Summarise team updates · Prepare meeting agendas | | **Analyst** | Summarise research reports · Draft executive summaries · Explain findings in plain language | | **Marketer** | Draft copy and content · Brainstorm campaign ideas · Adapt messages for different audiences | | **Engineer / Developer** | Generate boilerplate code · Debug and explain errors · Draft technical documentation | | **Operations** | Write SOPs and process documents · Create templates · Summarise policy documents | | **HR / L&D** | Draft job descriptions · Create training materials · Summarise candidate feedback | **One-line memory:** 👉 Every role has a **different entry point — same five modes underneath** ---

⚙️ Technical Deep Dive (Optional)

**What makes role-specific prompting different?** The five modes (Draft, Summarise, Research, Explain, Transform) apply universally — but the *context*, *tone*, *format*, and *audience* differ by role. A manager prompting for feedback: ``` "You are an experienced manager. Draft constructive feedback for a team member who consistently delivers quality work but misses deadlines. Tone: direct but supportive. Length: 3 paragraphs." ``` A developer prompting for documentation: ``` "Write technical documentation for this Python function. Target audience: other developers. Include: purpose, parameters, return value, example usage. Use docstring format." ``` Same mode (Draft). Completely different context, audience, and format. **How dots connect:** - Role-based context is an application of the Role prompting technique (from Effective Prompt Engineering) - The 4D framework's Description competency (from How to Think With AI) is exactly this — giving AI enough context to serve your specific role

--- # 🤝 5. The Human-AI Handoff — Who Does What ## 🧠 Core Intuition **Problem:** Without a clear model of who is responsible for what, teams either over-rely on AI or avoid it entirely. **What happens:** Think of every AI-assisted work task as a handoff — you own the start, AI handles the middle, you own the finish. ```mermaid flowchart LR A["🧑 YOU\nOwn the brief"] --> B["🎯 Clear task\nContext · Goal · Constraints"] B --> C["🤖 AI\nProduces the draft"] C --> D["📋 First output\nFast · Structured · Needs review"] D --> E["🧑 YOU\nReview + refine"] E --> F{"Good enough?"} F -->|"Yes"| G["📤 Final output\nYour name on it"] F -->|"No"| H["🔁 Re-prompt or edit"] H --> C style A stroke:#2980b9 style C stroke:#27ae60 style E stroke:#2980b9 style G stroke:#4a90e2 style H stroke:#f39c12 ``` **The handoff principle:** | Stage | Owner | What it looks like | |---|---|---| | **Define the task** | You | Write a clear, specific prompt with context | | **Produce the draft** | AI | Fast first version — never the final version | | **Review the draft** | You | Check accuracy, tone, completeness | | **Refine or re-prompt** | You + AI | Iterate until it's right | | **Approve the output** | You | Your name goes on it — your responsibility | **Why it matters:** The handoff model eliminates the most common workplace AI mistake — treating AI output as finished work. Every output is a draft until you approve it. **One-line memory:** 👉 You own the brief, AI owns the draft, you own the finish ---

⚙️ Technical Deep Dive (Optional)

**What is it called?** 👉 **Human-in-the-loop (HITL)** — a system design principle where human oversight is embedded at key stages of an AI workflow **Why this matters for professional accountability:** - AI systems have no legal or professional accountability — you do - In most professional contexts (legal, financial, medical, HR), AI output that you submit is treated as *your* output - The handoff model makes this explicit: AI is a tool, not an author **How dots connect:** - Diligence (from How to Think With AI) is the 4D competency that covers exactly this — checking, verifying, and taking responsibility - Hallucination (from How LLMs Work) is the architectural reason review is non-negotiable, not optional

--- # ⚠️ 6. Common Mistakes — What Goes Wrong at Work ## 🧠 Core Intuition **Problem:** Most workplace AI failures follow predictable patterns. Knowing them in advance is the fastest way to avoid them. **What happens:** These are the six most common mistakes people make when using AI at work. ```mermaid flowchart TD M["⚠️ Common AI Mistakes at Work"] --> M1["1. Treating output as final\nNo review before submitting"] M --> M2["2. Vague prompts\n'Help me with this report'"] M --> M3["3. Wrong task fit\nUsing AI for low-fit tasks"] M --> M4["4. No context given\nAI doesn't know your role, audience, or goal"] M --> M5["5. Pasting sensitive data\nSharing confidential info into public AI tools"] M --> M6["6. Expecting perfection\nFirst draft frustration → abandoning AI entirely"] style M stroke:#c0392b style M1 stroke:#c0392b style M2 stroke:#e67e22 style M3 stroke:#e67e22 style M4 stroke:#f39c12 style M5 stroke:#c0392b style M6 stroke:#f39c12 ``` **The fix for each:** | Mistake | Fix | |---|---| | Treating output as final | Always review before submitting — no exceptions | | Vague prompts | Use the 5-element prompt structure: Task + Context + Role + Format + Constraints | | Wrong task fit | Check the task-fit map before prompting | | No context given | Always tell AI your role, audience, goal, and any relevant constraints | | Pasting sensitive data | Know your organisation's AI data policy before using any AI tool | | Expecting perfection | First draft is the starting point — iteration is the process | **One-line memory:** 👉 The biggest workplace AI mistake: **submitting AI output you haven't reviewed** ---

⚙️ Technical Deep Dive (Optional)

**The data sensitivity problem — why it matters:** Most popular AI tools (ChatGPT, Claude, Gemini) are consumer or cloud products. When you paste content into them: - The data leaves your organisation's control - Depending on the tool and settings, it may be used for training - Many enterprise agreements prohibit this for confidential data **What to do:** - Know whether your organisation has enterprise agreements with AI vendors (which include data privacy protections) - Use approved internal AI tools for anything confidential - When in doubt: remove or anonymise sensitive details before prompting **How dots connect:** - Diligence (from How to Think With AI) covers data responsibility explicitly - Responsible AI (next module) addresses organisational AI governance in depth

--- # 🏗️ 7. Your AI Workflow — Building a Personal System ## 🧠 Core Intuition **Problem:** Individual AI use cases are useful. But the people who get the most from AI at work are those who build a **system** — not just a habit of occasionally asking AI for help. **What happens:** A personal AI workflow has three layers. ```mermaid flowchart TD W["🏗️ Your Personal AI Workflow"] --> L1["Layer 1: Trigger List\nWhich tasks will I always try AI for first?"] W --> L2["Layer 2: Prompt Library\nWhich prompts work well for my common tasks?"] W --> L3["Layer 3: Review Habit\nHow do I check AI output before using it?"] L1 --> EX1["Example: Every email draft over 3 paragraphs\nEvery meeting summary\nEvery first report draft"] L2 --> EX2["Example: Saved prompts for weekly reports,\nstakeholder emails, and project briefs"] L3 --> EX3["Example: Read once for accuracy,\nonce for tone, once for completeness"] style W stroke:#2980b9 style L1 stroke:#27ae60 style L2 stroke:#f39c12 style L3 stroke:#8e44ad ``` **Build your workflow in three steps:** **Step 1 — Your Trigger List** Write down 5 tasks you do every week that are good AI fits. These become your default "try AI first" list. **Step 2 — Your Prompt Library** For each trigger task, save a prompt that works well. Refine it over time. A library of 10 good prompts saves more time than a hundred one-off experiments. **Step 3 — Your Review Habit** Define how you review AI output for each task type. High-stakes output (going to clients, executives) gets full review. Internal drafts get a lighter pass. **One-line memory:** 👉 A personal AI workflow = **Trigger List + Prompt Library + Review Habit** ---

⚙️ Technical Deep Dive (Optional)

**What makes a good prompt library entry?** Each saved prompt should capture: - **Task type** (what mode: Draft / Summarise / etc.) - **Role instruction** ("You are a senior business analyst...") - **Context variables** (placeholders for the specific content: `[INSERT MEETING NOTES HERE]`) - **Format instruction** ("Respond in 3 bullet points, each under 20 words") - **Constraint** ("Do not include speculation. Only summarise what was explicitly said.") **Example saved prompt — meeting summary:** ``` You are a professional executive assistant. Summarise the following meeting notes into: 1. A 2-sentence overview of what was discussed 2. Key decisions made (bullet list) 3. Action items with owner and deadline (if stated) Tone: professional and neutral. Do not infer or speculate. [INSERT MEETING NOTES HERE] ``` **How dots connect:** - This is an application of the System Prompt and Constraint techniques (from Effective Prompt Engineering) - Delegation (from How to Think With AI) is the competency you are systematising here

--- # 🧠 Final Mental Model ## 🧠 Simple Version (For Everyone) ```mermaid flowchart TD TASK["📋 Work Task Arrives"] TASK --> FIT{"✅ AI fit?"} FIT -->|"High fit"| PROMPT["🎯 Write a clear prompt\nTask + Context + Role + Format"] FIT -->|"Medium fit"| CAREFUL["⚠️ Prompt carefully\nPlan to verify output"] FIT -->|"Low fit"| SKIP["🧑 Do it yourself\nAI can support, not decide"] PROMPT --> DRAFT["🤖 AI produces first draft"] CAREFUL --> DRAFT DRAFT --> REVIEW["🧑 You review:\nAccurate? Complete? On-brand?"] REVIEW -->|"Good"| OUTPUT["📤 Final output — your name on it"] REVIEW -->|"Needs work"| ITERATE["🔁 Refine prompt or edit draft"] ITERATE --> DRAFT style TASK stroke:#2980b9 style FIT stroke:#f39c12 style PROMPT stroke:#27ae60 style DRAFT stroke:#27ae60 style REVIEW stroke:#2980b9 style OUTPUT stroke:#4a90e2 style SKIP stroke:#c0392b ``` 👉 AI at work is not about trusting AI more. It is about using AI smarter. 👉 Your expertise does not go away. It becomes the quality filter on everything AI produces. 👉 The best AI users are not the fastest — they are the most deliberate. ---

⚙️ Technical Flow (For Professionals)

```mermaid flowchart TD subgraph YOU["🧑 Human Layer"] DEFINE["Define task + mode\n(Draft / Summarise / Research / Explain / Transform)"] CONTEXT["Add role, audience, format, constraints"] REVIEW2["Review for accuracy, tone, completeness"] APPROVE["Approve + take accountability"] end subgraph AI["🤖 AI Layer"] GENERATE["Generate first draft\n(pattern prediction at scale)"] ITERATE2["Iterate on refinement prompts"] end DEFINE --> CONTEXT --> GENERATE --> REVIEW2 REVIEW2 -->|"Iterate"| ITERATE2 --> REVIEW2 REVIEW2 -->|"Approve"| APPROVE subgraph SYSTEM["🏗️ Personal Workflow System"] TRIGGERS["Trigger List — when to use AI"] LIBRARY["Prompt Library — saved, refined prompts"] HABIT["Review Habit — how to check output"] end SYSTEM -.->|"Guides every task"| YOU style YOU stroke:#2980b9 style AI stroke:#27ae60 style SYSTEM stroke:#8e44ad ```

--- # 🚀 If You Remember Only Three Things 1. **AI at work is a colleague, not a vending machine.** Brief it well, review its work, take responsibility for the output. 2. **Five modes cover almost every use case:** Draft · Summarise · Research · Explain · Transform. Name the mode before you prompt. 3. **Build a system, not just a habit.** Trigger list + prompt library + review habit = consistent results at scale. --- # 📚 Concept Reference Card | Concept | One-Line Memory | Where it Applies | |---|---|---| | AI as colleague | Fast, capable, always needs review | Every AI-assisted task | | Draft mode | Let AI write the first version | Emails, reports, proposals | | Summarise mode | Compress content to key points | Meetings, documents, research | | Research mode | Surface options and background | Planning, strategy, analysis | | Explain mode | Translate complexity to clarity | Stakeholder comms, training | | Transform mode | Convert content from one form to another | Slides ↔ prose, bullets ↔ narrative | | Task fit | High / Medium / Low — match before prompting | Every task | | Human-in-the-loop | You own the brief and the finish | Consequential outputs | | Prompt library | Saved, refined prompts for recurring tasks | Workflow efficiency | | Review habit | How you check output before submitting | All final outputs | | Data sensitivity | Know your org's AI policy before pasting | Any confidential content | --- # 🎯 Success Metrics Track these to know if your team is genuinely using AI effectively at work: - ✅ Can name the five AI work modes and give a personal example of each - ✅ Uses the task-fit map before prompting — not just when something goes wrong - ✅ Never submits AI output without review - ✅ Has a personal trigger list of recurring tasks where AI is the first tool - ✅ Has at least 3 saved, refined prompts for their most common work tasks - ✅ Knows their organisation's AI data policy and applies it consistently - ✅ Treats AI output as a draft, not a deliverable --- # 📊 Evidence from Practice **Real example: Enterprise AI Adoption Programme** **Challenge:** - Employees understood what AI was but weren't using it consistently - Those who were using it had no review habits — submitting AI output unchecked - Managers concerned about quality and data confidentiality - No shared vocabulary for talking about AI use across teams - High variation in AI benefit — some people getting 2 hours/day back, others getting nothing **Approach:** - Half-day workshop on AI in the workplace for cross-functional teams - Emphasis on the colleague mental model — reframed what AI is and isn't - Task-fit mapping exercise: teams mapped their own recurring work tasks to fit levels - Prompt library sprint: each person left with 5 saved prompts for their role - Review habit design: teams agreed on minimum review standards by output type **Results:** - Immediate increase in consistent AI use across previously hesitant teams - Reduction in unchecked AI output being submitted — review habit adopted - Teams built shared prompt libraries that compounded in value over time - Managers reported higher confidence in AI-assisted work quality - Clear organisational language emerged for discussing AI tasks, fit, and accountability --- # 🎯 Book Your Workshop Ready to help your team move from knowing about AI to using it every day — confidently and responsibly? → **[Book a Discovery Call](https://calendly.com/naveen-innoventurex/free-consultation-with-naveen)** or → **[Contact me directly](/contact)** *** *Half-day workshop includes:* - *Interactive sessions on all 7 core concepts* - *Task-fit mapping exercise using your team's real work tasks* - *Role-specific use case deep dive* - *Prompt library sprint — leave with 5 saved prompts* - *Review habit design — agree minimum standards as a team* - *Personal AI workflow map as a takeaway*