PRESENTS · MAVEN COHORT 2 · SESSION 1 OF 5 SAT MAY 16, 2026 · 12:00 PM EDT AI for People Leaders From tinkering with AI to leading AI adoption. 5 live sessions · May 16 to June 20, 2026 · on Maven CO-INSTRUCTORS S Sandhiya Thiruvengadam Leader, HRBP & AI · Apexon Q Q Hamirani Chief People Officer, Highlevel · Adjunct Faculty, LBS Session 1 · Slide 1/15
What you told us before we started Three patterns rang loud across your responses. " I want to actually DO things, not just understand AI conceptually. To build, automate, and walk back in as someone who can credibly guide my team. A PEOPLE LEADER · COHORT 2 " I want to shift from thinking about AI as a personal productivity tool to thinking about it as an operating-model lever for HR. A PEOPLE LEADER · COHORT 2 " My biggest barrier? I don't know where to start. (You're not alone. That's the whole job of this cohort.) A PEOPLE LEADER · COHORT 2 THE THREE THEMES WE HEARD Build, don't just learn Hands on the keyboard, not slides. Operating-model thinking AI as a lever for the function. A clear place to start A first 3 Quick Wins, not a 50-tool list. We read every response. The next 6 weeks are built to answer these three statements. ↩ HOME Session 1 · Slide 2/15
Today's path 90 minutes · 7 segments · 1 module outcome ★ HOW TO TELL US YOU'RE LOST · NO JUDGMENT Type SLOW in chat any time, or raise your hand. We pause. We back up. The cohort is senior · the room moves at the slowest person's pace. AGENDA 12:00 Welcome + what you told us 12:10 The acceleration + three doors of AI 12:25 Vocabulary + the HR AI Automation Landscape 12:40 AI-Native journey + IMPACT framework 12:55 Which tool wins for which HR task 1:05 🤖 Use AI Now: plot your 3 quick wins 1:20 Skill anatomy + this week + reflection ★ TODAY'S NORTH STAR · BY 1:30 YOU'LL SAY "I can see exactly where AI will create the biggest impact in my organization, and how to prioritize strategically." One module. One outcome. Every minute of this session is in service of the line above. Module 1 outcome · keep it in mind ↩ HOME Session 1 · Slide 3/15
The acceleration: from 75 years to live now The technology curve isn't accelerating. It's vertical. 01 Telephone 75 years to 100M users 02 Smartphone 16 years 03 Internet 7 years 04 ChatGPT 3 months 05 Cowork live now VERIFIED · OCT 2025 TO MAR 2026 (6 MONTHS) 4 new Claude models 20+ dated product updates 1 net-new product · Cowork Source: support.claude.com release notes · 🟢 verified May 2026 WHY THIS MATTERS FOR HR 6 months, not 6 years, to adapt. ★ This is why Cohort 2 exists. ↩ HOME Session 1 · Slide 4/15
How to use Claude · the restaurant 9 ways in. One brain behind them all. We walk through every door across 5 sessions. FRONT OF HOUSE · what the guest sees YOU ARE HERE ① WALK IN Chat claude.ai · type back-and-forth Universal entry. Where everyone starts. Most people stay here. ② SIT AT A TABLE Claude Project a contained workspace with memory One Project per recurring task. Met in Session 2. ③ THE FILE ON YOU CLAUDE.md the profile that personalizes Your role, voice, rules · loaded every session, automatically. ④ THE MENU Skills markdown packs Claude can call Anatomy on Slide 12 today. First install: this week. ⑤ STANDING ORDERS Routines scheduled actions Claude runs for you "Every weekday 7AM: brief me." Built in Session 3. ⑥ THE RUNNER Dispatch parallel sub-agents on one job "Research 5 candidates at once." Met in Session 4. ⑦ KITCHEN LINE Computer Use Claude drives a browser for you "Pull these 20 PDFs and tag them." Met in Session 4. ⑧ THE RECEIPT Hooks what fires after every meal "After every reply, log to Notion." Mentioned in Session 4. ⑨ BACK OF HOUSE · the supply chain (drawn large because it carries everything) APIs · Connectors · MCP Where your data actually lives. The kitchen pulls from here. Three different things. We unpack the difference in Session 4. API Connector MCP Same kitchen, three ways to ask. → NEXT SLIDE · TRANSLATING If you remember the "three doors" framing from earlier training, the next slide shows how those map to the nine zones. ↩ HOME Session 1 · Slide 5/15
Same brain, three doors Every tool has a name for each. We teach the doors. Restaurant zones nest inside. 01 DOOR 1 · JUST TALK Chat Type back-and-forth. CALLED: Chat in Claude · ChatGPT · Gemini · Copilot All four call it the same thing. Universal entry point. RESTAURANT ZONES INSIDE · 1 ① WALK IN Chat ★ TODAY · WE STAY IN DOOR 1 02 DOOR 2 · PERSISTENT HELPER Cowork FILES Reads your files. Sits at your desk. CALLED: Cowork + Projects in Claude → Projects + Custom GPTs in ChatGPT → Gems + Workspace in Gemini · Copilot for M365 RESTAURANT ZONES INSIDE · 3 ② SIT AT A TABLE Claude Project ③ THE FILE ON YOU CLAUDE.md ④ THE MENU Skills ▷ OPENED IN SESSION 2 03 DOOR 3 · AUTONOMOUS BUILDER Claude Code $ claude › generate quarterly comp review reading: comp-philosophy.md reading: level-taxonomy.md drafting: q1-comp-review.docx ✓ ready for review $ _ Multi-step work while you sleep. CALLED: Claude Code + Computer in Claude → Codex CLI + Atlas in ChatGPT → Gemini CLI + Mariner · Copilot Agent + Studio RESTAURANT ZONES INSIDE · 5 ⑤ STANDING ORDERS Routines ⑥ THE RUNNER Dispatch ⑦ KITCHEN LINE Computer Use ⑧ THE RECEIPT Hooks ⑨ BACK OF HOUSE APIs · Connectors · MCP Product names change quarterly. The doors don't. The zones don't. Also Door 3: Perplexity Comet · Browserbase · Adept ACT-2. ↩ HOME Session 1 · Slide 6/15
The 8 words that unlock the rest Print this. Pin it. By Week 6 you won't need it. LLM Predicts next word. Powers every chatbot. Claude, ChatGPT, Gemini are all LLMs. RAG AI pulls from your trusted source first. Build a bot that answers only from your handbook. MCP Open standard for AI to talk to your tools. Granola, Gmail, Slack, Notion → all via MCP. Skill A reusable instruction file. Build once, run forever. Like a macro for AI. We build one in Week 2. Agent AI that takes multi-step actions on its own. Q's senior-leadership ATS. Claude Computer browsing. Cowork Claude reads your local files. Works alongside. Virtual assistant sitting at your desk. Markdown Plain text with # for headings, | for tables. By Week 6, less Excel and more markdown. CLAUDE.md The "About me" file Claude reads every session. The most important file you'll ever write. THE THROUGH-LINE AI + Human > AI alone. The vocabulary is just the floor. Your judgment is the ceiling. Every artifact in this cohort is co-authored with you. (BCG · David Green, 2025) ↩ HOME Session 1 · Slide 7/15
How Agentic AI has progressed Same recruiting task, four ways to run it. Each stage adds more agency than the last. The skill is knowing which one a problem belongs to. DETERMINISTIC INFERENCE AGENTIC AUTONOMOUS Workflow DETERMINISTIC · FIXED STEPS New Applicant ↓ Review Resume ↓ Schedule Interview ↓ Add to ATS YOU TODAY Every step is a human. The system is your team's time. AI Workflow INFERENCE · AI HELPS ONE STEP New Applicant ↓ ★ Screen Resume ↓ Schedule Interview ↓ Add to ATS SESSION 1 STARTS HERE You hand AI the slowest step. Save 30 minutes per resume. Agentic Workflow AI ORCHESTRATES · YOU APPROVE New Applicant ↓ ★ Research Candidate ↓ ★ Analyze Skills Match ↓ Add to ATS ↓ ★ Draft Email Teams Ping SESSIONS 2 + 3 LIVE HERE You design the pipeline. AI runs each step. You approve the gates. NEW Agent AUTONOMOUS · PURSUES THE GOAL New Applicant Employee Query ↓ ★ Intelligent Assistant Context & Instructions Knowledge Repository Action Framework ↓ ↓ Send Email Update HRIS SESSIONS 4 + 5 · CAPSTONE You commission the agent. It pursues the goal across many inputs and tools. THE COHORT TAKEAWAY · PROGRESSION FRAME via ZAPIER PEOPLE OPS · ADAPTED FOR HR Each stage adds more agency than the last. The skill is knowing which one a problem belongs to. You'll ship a Skill, a Routine, and an Agent by Week 6. ↩ HOME Session 1 · Slide 8/15
The AI-Native journey Where you are. Where you're going. Where your org needs to be. 1 START HERE AI-Curious YOU Aware AI exists. Occasional use. ORG Tools layered on top of existing processes. 2 WHERE MOST ARE AI-Capable YOU Regular use. Strong prompting. ORG Multiple teams using AI. Wins documented. 3 GOAL BY WEEK 6 AI-Native YOU AI is your default. You build Skills. ORG Processes redesigned with AI at the center. 4 THE BENCHMARK Frontier Org YOU Rebuilding the function. Others learn. ORG Operating-model shift. (e.g., Barona OS) The cohort's job: move you from 1 or 2 to 3 in six weeks. Stage 4 is the post-cohort horizon. STAGES 1 + 2 · WELL-TRODDEN TERRITORY If you're already at Stage 2, the cohort fast-forwards you. Most Cohort 2 participants already use Claude weekly. Stages 3 and 4 are where the actual work of this cohort happens. 3 UNPACKING STAGE 3 AI-Native HR Leader AI is your operating default, as fluent as Excel once was. Reach for Claude before you reach for the keyboard. 4 UNPACKING STAGE 4 Frontier Organization Core HR processes redesigned around AI, not bolted on. e.g., Barona OS (Finnish HR firm rebuilding the function on AI agents). ↩ HOME Session 1 · Slide 9/15
The IMPACT Framework How to find your first 3 AI quick wins, and ship them in 72 hours. I Identify Pinpoint HR processes where AI adds clear value M Map Match AI capabilities to specific HR needs P Pilot Run a small-scale, low-risk test first A Adopt Roll out to the team with training + change mgmt C Capacitate Build the team's AI muscle so it sticks T Track Measure outcomes. Iterate. Defend at QBR. WORKED EXAMPLE · COHORT 1 · 4 WEEKS "AI-assisted exit interview synthesizer." 6 hours saved per HRBP per quarter. CHRO approval on Week 5. Rolling out across BU's. WHERE EACH IMPACT LETTER LIVES IN THIS STORY I exit interview pain · M GenAI = summarization · P one BU pilot A manager training added · C HRBP-built Skill · T $ saved + retention lift to QBR SHIPS WITH WEEK 1 · COMPANION WORKBOOK HR_AI_Roadmap A draft worksheet that maps the 6 IMPACT steps to your own quick wins. 📥 v1 lands in cohort chat by Monday 🛠 We'll refine it together over the 5 weeks 📚 Inspired by David Green's Agentic AI Chart (MyHRfuture) (in the saved-posts library, Sandhiya's curation) ↩ HOME Session 1 · Slide 10/15
Which tool wins for which HR task Don't memorize. Bookmark. The most powerful AI move in HR is using the right tool for the job. Drafting JDs, policies, comms Strongest writing voice + handles long documents in one go Claude Image generation (org charts, training graphics) Claude doesn't generate images natively Gemini / ChatGPT Web search + citations Faster search + better source-citing UX Perplexity / ChatGPT Excel work inside Microsoft ecosystem Native M365 integration Copilot Building reusable Skills SKILL.md ecosystem is most mature Claude Cross-platform (Slack · Notion · Gmail) MCP connectors. Biggest competitor gap. Claude (via MCP) Sensitive HR data work (PII · comp · medical) Best privacy guarantees + clearest data residency. Always your enterprise tool first. Claude OR your enterprise tool This cohort teaches with Claude. The doors translate. The decisions stay yours. ↩ HOME Session 1 · Slide 11/15
🤖 Use AI Now Take 5 minutes. Plot YOUR first 3 Quick Wins. The picking is harder than the generating. PASTE THIS INTO CLAUDE Copy ⌘C I work in [your HR role] at a [size] company in [industry]. Looking at HR functions: Recruiting, EX, L&D, Performance. Generate 10 specific AI use cases for my function. Each labeled with: • Complexity (Low/Med/High) • Business impact (Low/Med/High) Format as a markdown table. ① ROLE + CONTEXT ② SCOPE LOCKED ③ QUANTITY ANCHOR ④ STRUCTURED OUTPUT ⑤ FORMAT SPEC 5 THINGS THIS PROMPT DOES RIGHT NEW WORD ↑ Markdown = plain text with simple symbols (# for headings, | for tables). Claude reads & writes it natively. Slide 6. THEN PLOT YOUR TOP 3 HERE B U S I N E S S I M P A C T ↑ IMPLEMENTATION COMPLEXITY → QUICK WINS Low complexity · High impact YOU PICK 3 FROM HERE 1 2 3 STRATEGIC High complexity · High impact Park these for Project 2 EXPERIMENTS Low complexity · Low impact OK to tinker · don't ship AVOID ZONE High complexity · Low impact Drop. Move on. ▸ MAKE THIS PROMPT EVEN BETTER · THREE MOVES + ADD CONTEXT · FILES Attach your last 3 JDs, your engagement survey, your org chart. Claude reads it all. ? INTERVIEW ME FIRST "Before you generate, ask me 10 questions to surface non-obvious use cases." ⌗ SAVE TO A PROJECT Pin the output to a Claude Project so Claude remembers your context next time. ↩ HOME Session 1 · Slide 12/15
Anatomy of a Skill What you'll see when you open one. Three named parts. Install one this week. SKILL.md · hr-job-description-writer 3 PARTS THIS SKILL HAS 1 --- 2 name: hr-job-description-writer 3 description: Drafts JDs in your 4 company voice and level taxonomy. 5 --- 6 7 # When invoked 8 When the user asks for a job description: 9 10 1. Ask for role title, level, and team. 11 2. Pull tone reference from /examples. 12 3. Output structure: 13 Overview → Responsibilities 14 → Requirements → Comp band 15 16 4. Flag any compensation language 17 for [LEGAL REVIEW]. ⚙ NAME · what Claude calls it ⚙ DESCRIPTION · when to reach for it ⚙ BODY · the instructions Claude follows YOUR WEEK 1 INSTALL · NO TERMINAL ① claude.ai → New Project. ② Paste a SKILL.md into Project knowledge. ③ Start chatting. Claude now follows the Skill. NEW WORD ↑ SKILL A Skill is a reusable markdown recipe Claude follows. By Sat May 23: 1 Skill installed · 1 CLAUDE.md draft · 1 Quick Win committed. ★ HOW THIS TRANSLATES TO YOUR STACK same idea, different brand · pick the door your company already pays for ★ CLAUDE · RECOMMENDED Skills (SKILL.md) MICROSOFT COPILOT Copilot Studio agent · Power Automate templates GOOGLE GEMINI Gems · Custom instructions OPENAI CHATGPT Custom GPTs · GPT Instructions ↩ HOME Session 1 · Slide 13/15
Take these with you Bookmark. Reference. Reuse. Plus where you are in the 5-week arc. RESOURCE PACK · CITED TODAY 📊 WORKBOOK · IN MAVEN HOME HR_AI_Roadmap v1 10-tab IMPACT workbook with 12 pre-filled HR pain points. Companion to Slide 9 today. Use this for Project 1.1. FILE · HR_AI_Roadmap_v1.xlsx 🧩 SKILLS BUILD KIT · FILTERABLE HTML 20 HR Skills · build your own 20 HR-specific Skills with full SKILL.md specs + EXAMPLES. Each card = one spec + one COPY button. Build for your org, not generic. PAGE · skills-build-kit.html 🛠 SETUP GUIDE · FOR SESSION 2 The Builder Path 5-step walkthrough: GitHub account, terminal basics, Claude Code install. Do this between Session 1 and Session 2. We'll use it together May 30. PAGE · builder-path.html 🤖 REFERENCE PROMPT · FOR SESSION 2 Sandhiya's CoS prompt Her actual Microsoft Copilot CoS template: 8-section briefing, evening reflection. Change the [BRACKETS]. Port it to Claude. That's Project 2.1. FILE · 07_Shareable_Prompt_Template.md 🌐 EXTERNAL · ANTHROPIC GITHUB Anthropic Skills marketplace Official Skills repo. Browse, fork, install. New skills added weekly. Bookmark this. By Session 3 you'll be reading SKILL.md files like English. URL · github.com/anthropics/skills 📋 EXTERNAL · INFOGRAPHIC "How to use Claude" Ruben Hassid's 9-capability ladder with "try this first" prompts. A complementary view of the same restaurant · flat ladder instead of zones. URL · how-to-ai.guide WHERE YOU ARE · 5-WEEK ARC TODAY SESSION 1 · MAY 16 Start where you are The landscape. The 8 words. Anatomy of a Skill. ✓ You shipped: Skill installed. SESSION 2 · MAY 30 Build the AI for you Chief of Staff. CLAUDE.md. Builder Path setup. → You ship: CoS v1. SESSION 3 · JUN 6 Make it safe to ship Sandhiya's CoS + Q's ATS. Routines + Ethics + Code. → You ship: 1 Routine running. SESSION 4 · JUN 13 Deploy + make the case MCP. Connectors. Dispatch. Shark Tank Round 1. → You ship: MCP installed. SESSION 5 · JUN 20 Capstone showcase Live demos. Peer awards. Graduation. → You ship: the full thing. SAVE THIS SLIDE Everything Q and I will reference across the cohort, in one place. We add to it each week. ↩ HOME Session 1 · Slide 14/15
This week Three things happen between now and Session 2 (Saturday, May 30). 1 DELIVERABLE · DUE FRI MAY 22 Project 1.1 Strategic AI Assessment ~60 minutes of work 📊 Org AI readiness 300-400 words 📍 Opportunity matrix 8+ items plotted on the 2x2 ⚡ 72-hour action plan For your top 3 Quick Wins 🎤 3-min summary ready You'll present in Session 2 2 FIELD GUIDE · DAILY · ~10 MIN Week 1 reps SUN Install your first HR Skill 5 min · from the marketplace doc MON Run it on a real task 10 min · screenshot the output TUE Draft your CLAUDE.md 15 min · template provided WED Open Door 2 (Cowork) 15 min · point at a folder you own THU Pick your Quick Win #1 10 min · commit publicly in cohort chat FRI Ship Project 1.1 ~60 min · roadmap + matrix + plan 3 SUNDAY · COHORT CHAT 3 reflection prompts 1 What surprised you most about GenAI's HR potential? 2 How might it transform YOUR specific role? 3 What's one Quick Win you'll commit to before Session 2? NEXT TIME See you Saturday, May 30, 12:00 PM EDT for Session 2. Bring your Quick Win. ↩ HOME Session 1 · Slide 15/15
End of Session 1. You shipped. Now keep the muscle warm. ★ NEXT SESSION Session 2 · Build the AI that works like you Saturday May 30, 2026 · 12:00 PM EDT · live on Maven OPEN DECK → Between now and then · 10 minutes a day on Atlas 30. Ship one thing. ↩ HOME End of Session 1
SESSION 1 DECK · COHORT 2

Cohort 2 is in the room.Session 1 — AI Strategy & Your First Use Case. These slides walk through it live with Q & Sandhiya. Cohort members sign in to open the deck; everyone else: catch Cohort 3.

ALREADY IN COHORT 2?

Welcome back. The deck loads behind the sign-in.

WATCHING FROM OUTSIDE?

Cohort 3 opens later this summer. We'll save you a seat.

CATCH COHORT 3 →
🏠 1 / 12