Claude AI Classroom Training in Mumbai
| Claude AI Content | ||
|
MODULE 1 - Getting Started with Claude : 1.1 - What is Claude : |
||
|
~ Claude as a large language model — what that means in plain terms ~ How Claude is different from ChatGPT, Gemini, and Copilot ~ Claude’s strengths — long documents, nuanced reasoning, writing quality, instruction following ~ Limitations — no live internet by default, knowledge cutoff, hallucination risk |
||
| 1.2 - The Chat Interface : | ||
|
~ Navigating claude.ai — conversations, sidebar, model selector ~ Starting a conversation and understanding the turn-based model ~ Uploading files — PDFs, images, spreadsheets, documents ~ Editing and regenerating responses ~ Free vs Pro vs Team vs Enterprise plans — what changes at each level |
||
| 1.3 - Tokens and Context : | ||
|
~ What a token is — roughly 4 characters or 0.75 words ~ Input tokens vs output tokens — how both count toward usage ~ The context window — what Claude can hold in a single conversation ~ Why long conversations degrade quality and what to do about it ~ Token consumption for common tasks — short email vs long report vs document analysis ~ Strategies for staying within limits — summarise, segment, start fresh |
||
| Use Cases — Chat and Basic Interaction : | ||
|
~ Summarise a lengthy report or research paper into key points ~ Ask Claude to explain a complex policy, regulation, or technical document in simple language ~ Get a quick answer or definition without switching to a search engine ~ Use Claude to check a document for errors, inconsistencies, or gaps |
||
|
MODULE 2 - Writing Effective Prompts : 2.1 - Why Prompting Matters : |
||
|
~ The garbage-in-garbage-out principle applied to AI ~ How Claude interprets a prompt — what it looks for and what it fills in by assumption |
||
| 2.2 - Prompt Anatomy : | ||
|
~ Role — who Claude should be in this task ~ Goal — what output is expected ~ Context — background information Claude needs ~ Format — how the output should be structured ~ Constraints — what to avoid, tone, length, audience |
||
| 2.3 - Prompting Frameworks : | ||
|
~ GCA — Goal, Context, Action ~ CARE — Context, Action, Result, Example ~ DIG — Define, Instruct, Guide |
||
| 2.4 - Advanced Prompting Techniques : | ||
|
~ Chain-of-thought — asking Claude to reason step by step ~ Few-shot examples — showing Claude what good output looks like ~ Iterative refinement — treating the first output as a draft ~ System-level instructions — setting persistent context at the top of a conversation |
||
| Use Cases — Prompting | ||
|
~ Draft a professional email to a senior stakeholder explaining a delay ~ Write a job description for a specialist role using company values ~ Prepare five interview questions tailored to a specific competency framework ~ Analyse a competitor’s public report and identify strategic positioning ~ Rewrite a dense internal policy into plain language for frontline staff |
||
|
MODULE 3 - Projects : 3.1 - What Projects Are : |
||
|
~ Projects as persistent workspaces — Claude remembers instructions and context across sessions ~ How Projects differ from standard conversations ~ Availability — Pro and above plans |
||
| 3.2 - Setting Up a Project : | ||
|
~ Creating a Project and writing project-level instructions ~ Adding documents to the Project knowledge base — SOPs, tone guides, reference data ~ What Claude can and cannot access within a Project ~ Naming, organising, and sharing Projects with a team |
||
| 3.3 - Working Inside Projects : | ||
|
~ Running multiple conversations within the same Project ~ Updating Project instructions without breaking existing conversations ~ Best practices — one Project per client, campaign, or function |
||
| Use Cases — Projects : | ||
|
~ Set up a Project for a client account with brand guidelines, tone of voice, and key contacts embedded ~ Create a Project for a recurring process such as monthly reporting with reusable templates loaded in ~ Build a Project for HR onboarding with company policies and FAQs as the knowledge base ~ Use a Project for a product launch to keep all assets, briefs, and copy consistent across the team |
||
|
MODULE 4 - Artifacts : 4.1 - What Artifacts Are : |
||
|
~ Artifacts as distinct outputs — documents, code, HTML, SVG, React components, markdown ~ How Artifacts appear in the interface — side panel, live preview ~ When Claude chooses to create an Artifact vs respond inline |
||
| 4.2 - Types of Artifacts : | ||
|
~ Text documents — reports, policies, templates, proposals ~ Markdown — structured content with formatting ~ Code — Python, JavaScript, SQL, and other languages ~ HTML — web pages, interactive tools, styled layouts ~ Visual outputs — SVG diagrams, charts, infographics |
||
| 4.3 - Iterating on Artifacts : | ||
|
~ Asking Claude to revise an Artifact in the same conversation ~ Version control — switching between earlier and later versions ~ Copying, downloading, and sharing Artifact content |
||
| Use Cases — Artifacts : | ||
|
~ Generate a formatted project proposal as a document Artifact that can be downloaded directly ~ Build an interactive HTML calculator or decision tool for internal use ~ Create a data visualisation as an SVG chart from a pasted table ~ Write a SQL query or Python script as a standalone Code Artifact ~ Produce a styled one-pager or briefing document as an HTML Artifact |
||
|
MODULE 5 - Skills : 5.1 - What Skills Are : |
||
|
~ Skills as reusable instruction sets that give Claude specialist behaviour ~ How Skills differ from one-off prompts — they are persistent and repeatable ~ Where Skills live — attached to Projects or loaded at conversation start |
||
| 5.2 - Building a Skill : | ||
|
~ Defining the purpose and scope of a Skill ~ Writing the Skill instruction file — structure, triggers, output rules ~ Testing and refining a Skill before deploying to a team ~ Combining multiple Skills in a single Project |
||
| Use Cases — Skills : | ||
|
~ A writing style Skill that enforces company tone, vocabulary, and formatting rules across all communications ~ A data analysis Skill that applies a consistent framework — ABC classification, Pareto, trend analysis — to any dataset ~ A legal review Skill that checks contracts for common risk clauses and flags them for attention ~ A market intelligence Skill that structures competitor or industry inputs into a standard report template ~ A training design Skill that converts subject-matter content into learning objectives, activities, and assessments |
||
|
MODULE 6 - Cowork : 6.1 What Cowork Is |
||
|
~ Cowork as Claude’s agentic capability — taking multi-step actions on your behalf ~ The difference between chatting with Claude and delegating a task to Claude ~ How Cowork integrates with files, apps, and the browser |
||
| 6.2 - How Cowork Works : | ||
|
~ Writing a delegation brief — what to include for Claude to act without interruption ~ Monitoring a Cowork task — when Claude checks in vs proceeds independently ~ File access and output — where results are saved and how to review them ~ Cowork with browser access — Claude navigating web pages and extracting information |
||
| 6.3 - When to Use Cowork vs Chat : | ||
|
~ Use chat for exploratory, creative, or back-and-forth tasks ~ Use Cowork for tasks that have a defined output, involve file manipulation, or require multiple steps ~ Risk and review — always check Cowork outputs before acting on them |
||
| Use Cases — Cowork : | ||
|
~ Research a topic across multiple web sources and compile a structured briefing document ~ Process a folder of uploaded reports and produce a summary table for each one ~ Draft and format a complete slide deck from a provided outline, saving it as a downloadable file ~ Analyse a spreadsheet, apply a scoring model, and produce a ranked output file ~ Monitor a set of websites for updates and report changes in a daily digest |
||
|
MODULE 7 - Claude Design : 7.1 - What Claude Design Is : |
||
|
~ Claude Design as a visual creation capability — producing polished designs from a prompt ~ How Claude Design differs from image generation tools like DALL-E or Midjourney ~ Output types — slides, one-pagers, infographics, UI mockups, branded layouts |
||
| 7.2 - Working with Claude Design : | ||
|
~ Writing effective visual prompts — describing layout, hierarchy, tone, and colour ~ Providing brand inputs — logo, colours, fonts, reference designs ~ Iterating on a design — requesting changes and understanding what Claude can adjust ~ Exporting and using Claude Design outputs in other tools |
||
| Use Cases — Claude Design : | ||
|
~ Create a branded one-page summary of a project or proposal for executive review ~ Design a campaign or event announcement visual for internal or external communication ~ Produce an infographic from a data table or process description ~ Generate a slide template or presentation layout aligned to company visual identity ~ Mock up a UI screen or dashboard layout before handing off to a development team |
||
|
MODULE 8 - Putting It All Together : 8.1 - Choosing the Right Capability for the Task : |
||
|
~ Decision framework — when to use chat, Projects, Artifacts, Skills, Cowork, or Design ~ Combining capabilities — Projects with Skills and Artifacts, Cowork producing Artifacts |
||
| 8.2 - Building Your Personal Claude Workflow : | ||
|
~ Mapping your most time-consuming tasks to Claude capabilities ~ Designing a starter prompt library for your role ~ Setting up your first Project with the right Skills attached |
||
| 8.3 - Responsible Use and Quality Control : | ||
|
~ Always verify factual claims in Claude outputs before using or sharing ~ Data confidentiality — what not to put into Claude ~ Copyright, attribution, and originality considerations ~ Human judgement remains essential — Claude is a tool, not a decision-maker |
||
| Use Cases — Artifacts : | ||
|
~ Generate a formatted project proposal as a document Artifact that can be downloaded directly ~ Build an interactive HTML calculator or decision tool for internal use ~ Create a data visualisation as an SVG chart from a pasted table ~ Write a SQL query or Python script as a standalone Code Artifact ~ Produce a styled one-pager or briefing document as an HTML Artifact |
||
|
Online and personalised Claude AI training programmes are also available for teams and individuals. |
