GenAI Training on the Claude Stack.
A 3-day, immersive, compliance-safe AI bootcamp. Built on Claude Code, Cowork, Skills, and MCP.
Productivity. Governance. Competitive edge.
Three reasons Anthropic-curious enterprises are running structured cohorts now, not next year.
The depth (Code, Cowork, Skills, MCP) sits unused. Most teams stop at the browser.
Connectors put your CRM, Drive, and inbox inside Claude. Approval flows and hooks matter more than ever.
Without structured training
- Employees are already experimenting with Claude.
- Often in unstructured, inefficient, or risky ways.
- Hours wasted on trial-and-error. Compliance breaches creep in.
With structured training
- Save time on routine tasks. Hours → minutes.
- Move beyond chat to Claude Code, Skills, and MCP.
- Adopt AI safely, with hooks, approval flows, and audit.
How we will run the workshop.
Every cohort starts with a diagnostic survey, so the curriculum fits your team.
We survey every participant before Day 1 and rebuild the agenda around what we learn.
| Axis | What we ask | What it changes |
|---|---|---|
| 01. Skill baseline | Frequency of Claude use today. Tools tried (Code, Cowork, Skills). Confidence with data, dashboards. | Depth and pace of Day 1 foundations. Whether to bundle a Claude Code primer. |
| 02. Role-specific wishlist | Top 3 tasks to speed up. Workflow pain points by function. Real data they work with. | Day 3 Claude Code agent build uses their wishlist. Demos on analogues of their data. |
| 03. Our recommendation | Gaps the team has not flagged but will hit. Compliance and governance for their sector. | Adds compliance, hooks, approval-flow patterns where relevant. Sequences foundations so advanced users learn too. |
Output: a customised 3-day agenda shared with the sponsor 5 working days before Day 1.
Same 6 streams. Different weight, depth, examples.
Survey
10-minute online survey to every participant, 2 weeks before Day 1.
Diagnostic read-out
Sponsor gets skill distribution, top 5 use-case asks, our recommendations.
Agenda locked
Final 3-day agenda, tool list, exercise set, 5 working days before Day 1.
Delivery
Each day opens with the team's own examples. Compliance and Anthropic-stack governance woven in.
Result: participants learn on their own problems, not case studies. Our score is 9.11 / 10.
From foundational concepts to advanced application development.
Six streams. Tilted across the 3-day program based on the diagnostic.
01. Foundations & Briefing
Core prompting Claude. Above the line vs below the line AI.
02. Refresher in the Age of AI
How to understand and present data, enhanced with AI.
03. Building Apps. Basics.
App-thinking foundations before AI is layered in.
04. Claude for Dashboards
Claude in Excel and Sheets for daily reporting.
05. Claude Code Agents
The 2026 Anthropic builder stack, on real data.
06. MCP & Workflow Automation
From manual workflows to agents that run themselves.
Foundation, data, engagement, analytics, compliance, apps.
What they will be able to do.
Brief Claude with the precision your team's work deserves. Prompts, context, and Skills.
Use Claude in Excel and Sheets to analyse complex datasets in minutes, not hours.
Apply Claude to enterprise tools (CRMs, emails, reports) and turn raw data into strategic insight.
Use Claude Code, Cowork, and MCP to prototype and ship agents tailored for organisational use.
Three days. One arc. Real artifacts.
| Day | Theme | Focus areas |
|---|---|---|
| Day 01 | Foundations & Briefing Claude | Core concepts, prompting, problem framing, briefing well. |
| Day 02 | Data Storytelling, Userflows | How to understand and present data, enhanced with Claude. |
| Day 03 | Claude Code, Compliance & Apps | Anthropic-stack agent building, hooks and approval flows, cross-functional use cases. |
Setting the foundation. With Claude as anchor.
| Session | Topic | Format |
|---|---|---|
| 1.1 | Welcome, Icebreaker & Workshop Overview | Interactive intro |
| 1.2 | GenAI Fundamentals on the Claude Stack: What, Why & How | Talk + live demo |
| 1.3 | Prompt Engineering 101 (briefing Claude well) | Prompt Studio Lab |
| 1.4 | Hands-On Lab: Crafting Prompts for your use-cases | Group exercise |
| 1.5 | Real-World Prompt Use Cases (Summaries, Emails, Notes) | Demo showcase |
| 1.6 | Identify and solve problems worth solving | Video + exercise |
| 1.7 | Debrief & Day-End Challenge | Reflective quiz |
From data to narrative to first build.
| Session | Topic | Format |
|---|---|---|
| 2.1 | Recap + Warm-up Prompt Quiz | Interactive poll |
| 2.2 | Storytelling + Data Visualisation with Claude | Video + session |
| 2.3 | Data analysis & storytelling exercise | Hands-on exercise |
| 2.4 | Data visualisation de-brief | Discussion + demo |
| 2.5 | Intro to Claude Code, Cowork & MCP | Interactive session |
| 2.6 | Start building (Claude Code agent) | Hackathon |
| 2.7 | De-brief | Interactive |
Build, govern, ship.
| Session | Topic | Format |
|---|---|---|
| 3.1 | Introduction to advanced use-cases (Skills, MCP, Hooks) | Session |
| 3.2 | Continue building. Hands-on guidance with Claude Code | Break-out activities |
| 3.3 | Approvals, Audit & Hooks. Human-in-the-loop patterns | Talk + design exercise |
| 3.4 | Shark Tank format demo (internal contest) | Demo |
| 3.5 | What's Next: 30 / 60 / 90 Day Plan, Resources, Office Hours | Session |
| 3.6 | Closing ceremony and pledge-taking | Interactive exercise |
The Claude stack. Customised per cohort.
Indicative and not an exhaustive list.
Cross-stack track (optional, for hybrid cohorts): Microsoft Copilot, Power Automate. Used in selected modules where the cohort runs a mixed stack.
No lecture marathons.
100% interactive. Doing beats listening, every time.
- Hands-on exercises for doing, not just listening
- Kahoot and other interactive speed-quizzes
- Miro AI Idea Wall for breakout design sprints
- Real-time role-play
- Live builds in Claude Code with MCP wiring
Past trainings effectiveness.
Likelihood to recommend, out of 10.
Voice of participants and leaders from cohorts at a pharma and an energy company.
Rated 4 or 5 for overall satisfaction.
Voice of participants and leaders from cohorts at a pharma and an energy company.
Felt surprised at their own capabilities.
Voice of participants and leaders from cohorts at a pharma and an energy company.
Rated 4 or 5. 75% rated 5-star.
Voice of participants and leaders from cohorts at a pharma and an energy company.
Praise for Nimit's training.
"Very good and great trainer. Energetic, patient, and always willing to help. Everyone, from newbies to experienced participants, walked away with real value."
"Patient, with knowledge in abundance, and explained everything clearly. The session showed me how AI can reduce my manual workload and automate reporting."
"The energy and enthusiasm in all sessions were infectious. The dashboard exercise on Day 3 was a real wow moment."
Practical, hands-on, and immediately useful.
"Nimit helped both beginners and experienced participants learn at the same pace. Everyone ended up as a student who walked away with something new."
"This training made me realise I can create dashboards and reports in minutes instead of hours. It will transform how I approach my work."
"Practical, hands-on, and immediately useful. I've already started planning to automate my daily reports."
"Energetic sessions filled with real-world examples. I now know how to wire up Claude with my data to generate insightful reports."
From the sponsor's seat.
We recently had a 3-day Advanced AI training with Nimit Kathuria, and we were very impressed with the depth of knowledge and expertise he brought into the sessions. His content and subject command were truly valuable for the participants. It was a very positive learning experience, and we would be glad to explore more sessions with him in the future.
Enterprise Cohort Leader
Why it's more effective than other similar trainings.
They start with tools, not thinking.
They teach features, not frameworks. When you have a hammer, every problem looks like a nail.
What many trainers teach
- Prompts and features
- Tool demonstrations
- Technical capabilities
What teams actually need
- Problem framing
- Decision support
- Actionable outcomes
Business First. AI Second.
I don't start with AI. I start with your business problems. First we clarify on what is a problem worth solving. Only then do we bring AI into the conversation.
Identify what's worth solving
Half the session is here. The smartest question beats the fanciest prompt.
Translate business logic into AI logic
So the cohort can repeat the move on Monday, not just remember the demo.
Apply tools with clarity and impact
Claude Code, Cowork, Skills, MCP. The right tool for the right shape of problem.
Not an engineer. A practitioner.
A non-engineer who first understands the business problem and then builds Claude Code agents and MCP servers.
Our lead trainer isn't a traditional coder. He has built AI prototypes, Claude Code agents, custom MCP servers, dashboards, and automations as a marketing leader. This unique perspective allows him to explain complex technical ideas in plain language, making non-tech teams confident, not confused.
Energetic, patient, and always willing to help everyone.
Pharma Brand Participant
Nimit Kathuria. Practitioner first, educator second.
Nimit Kathuria
ISB Graduate. Visiting Faculty at Jai Hind College. Guest Lecturer at XLRI and SDA Bocconi. 15+ years in BFSI, Health, Consumer, Research, Advertising, and SaaS.
About Clearthought Academy.
Hands-on corporate learning. Real business outcomes.
We bridge the gap between theory and practice with practitioner-led training for enterprise teams.
Start your AI transformation.
Align goals and priorities.
15–20 employees. Run a 2 or 3-day program.
Productivity, adoption, and quality improvement.
Build an AI-ready, Claude-fluent workforce.
Ready to set up a meeting?
hello@clearthought.academy
+91 9250 107 515
Let's make your teams AI-confident.
Hands-on corporate learning. Real business outcomes. Built on the Claude stack you are already exploring.
For leaders, innovators, and advanced functions.
Programs we run alongside partners.
- Custom evals and observability for Claude agents
- Multi-agent orchestration patterns
- RAG at scale plus retrieval governance
- Anthropic SDK deep dives, fine-tuning workflows
- Voice and multimodal agents (with Claude)
- Production deployment and SRE for Claude-stack agents