These step-by-step tutorials demonstrate how to use Machine for complex, real-world tasks. Each tutorial provides detailed instructions, examples, and best practices.
This tutorial walks through using Machine to analyze a dataset, generate insights, and create visualizations.
1
Prepare Your Data
Start by uploading your dataset to Machine. Supported formats include CSV, Excel, JSON, and SQL databases.
You: "I need to analyze this customer feedback dataset"Machine: "I can help with that. Would you like to upload the dataset now?"
After uploading, Machine will automatically examine the data structure and provide an initial summary.
2
Clean and Transform
Next, clean your data to handle missing values, outliers, and inconsistencies.
You: "The dataset has some missing values and outliers we need to address"Machine: "I notice there are 23 missing values in the 'customer_age' column and 12 potential outliers in 'purchase_amount'. Would you like me to:1. Remove rows with missing values2. Impute missing values using the mean/median3. Handle these issues another way?"
Machine will suggest transformations based on data quality issues it detects, but you maintain control over how to proceed.
3
Exploratory Analysis
Have Machine perform initial analysis to understand patterns and relationships in the data.
You: "What are the main patterns you notice in this data?"Machine: "Based on my analysis, here are the key insights:- Customer satisfaction correlates strongly with response time (r=0.72)- The 25-34 age group represents 42% of your customer base- Product category A has the highest return rate at 8.3%Would you like me to generate visualizations for these insights?"
Machine can identify correlations, trends, clusters, and outliers automatically.
4
Visualize Results
Create visual representations of your data insights.
You: "Yes, please create visualizations for the key insights"Machine: "I'll create several visualizations to highlight these patterns. Which format would you prefer for the final report: PDF, interactive dashboard, or presentation slides?"
Machine will generate appropriate visualizations based on data types and relationships, including:
Bar charts for categorical comparisons
Scatter plots for correlations
Time series for trends
Heatmaps for complex relationships
5
Generate Recommendations
Finally, have Machine provide actionable recommendations based on the analysis.
You: "Based on this analysis, what actions should we consider taking?"Machine: "Based on the data, here are my recommendations:1. Reduce response times for the 25-34 demographic to improve satisfaction2. Investigate quality issues in Product Category A to address high return rates3. Expand inventory for high-margin products showing consistent growthWould you like me to prepare a detailed action plan for any of these recommendations?"
Machine can help prioritize actions based on potential impact and implementation difficulty.
This tutorial demonstrates how to use Machine to build a complete website from scratch.
1
Define Requirements
Start by describing your website needs and goals.
You: "I need to create a website for my photography business"Machine: "I'd be happy to help create your photography website. To get started, could you tell me:1. What's the primary purpose of the website? (portfolio, booking, sales)2. How many pages do you envision?3. Do you have any design preferences or examples of sites you like?"
Be as specific as possible about your requirements, but don’t worry if you don’t have all the details figured out—Machine will help fill in the gaps.
2
Design the Site Structure
Work with Machine to plan your website’s architecture.
You: "I need a portfolio showcase, about page, services/pricing, and contact form"Machine: "Based on your needs, here's a proposed site structure:- Home (with featured work and introduction)- Portfolio (organized by photography categories)- About (your story and approach)- Services & Pricing- ContactDoes this structure work for you, or would you like to make adjustments?"
Machine will suggest a logical organization based on your business type and website purpose.
3
Create Content
Develop the text and visual content for your website.
You: "I need help writing the content for these pages"Machine: "Let's start with the About page. Could you tell me a bit about your background, style, and approach to photography? I'll help craft that into engaging website copy."
For visual elements, you can:
Upload your own images
Have Machine suggest stock photos
Get guidance on creating a cohesive visual style
4
Build the Website
Let Machine generate the actual website code based on your approved design and content.
You: "Now let's build the actual website"Machine: "I'll create your website using modern, responsive design principles. Do you have a preference for:1. A static site (faster, simpler)2. A content management system like WordPress (more easily editable)3. A custom solution with specific features?"
Machine can generate:
HTML/CSS/JavaScript for static sites
WordPress themes and configurations
React/Next.js applications for more advanced requirements
5
Preview and Refine
Review the website and make iterative improvements.
You: "Can I see how the site looks so far?"Machine: "Here's a preview of your website. You can navigate between pages to see the current design and content. What aspects would you like to adjust?"
Machine will provide a live preview environment where you can:
Navigate between pages
Test responsive behavior for different screen sizes
Experiment with different design elements
6
Deploy Your Website
Finally, publish your website to make it accessible online.
You: "The site looks great! How do I publish it?"Machine: "I can help you deploy your website. Do you already have:1. A domain name2. Web hostingIf not, I can guide you through acquiring these and then help with the deployment process."
Machine can assist with:
Domain registration
Hosting setup
DNS configuration
Deployment to platforms like Netlify, Vercel, or traditional hosts
This tutorial shows how to use Machine to conduct comprehensive research on a topic.
1
Define Research Goals
Begin by specifying your research objectives and scope.
You: "I need to research the impact of artificial intelligence on healthcare"Machine: "I'll help you research AI in healthcare. To focus our efforts, could you specify:1. Any particular aspects of healthcare you're most interested in?2. Whether you need academic sources, news articles, or both?3. How recent the information should be?4. The approximate length or depth of research needed?"
Clear research parameters help Machine deliver more relevant results.
2
Gather Sources
Let Machine collect relevant information from multiple sources.
You: "Focus on diagnostic applications from the past 3 years, using both academic and news sources"Machine: "I'll gather information on AI in healthcare diagnostics from the past 3 years. This will include academic papers, news articles, and industry reports. This will take a few moments."
Machine will:
Search academic databases for relevant papers
Find news articles from reputable sources
Identify industry reports and analyses
Filter for recency and relevance
3
Analyze and Synthesize
Have Machine analyze the information and identify key themes and findings.
You: "What are the main findings from your research?"Machine: "Based on my analysis of 28 academic papers and 42 news articles, here are the key findings on AI in healthcare diagnostics:1. Diagnostic Accuracy: Studies show 15-30% improvement in early detection rates for conditions like diabetic retinopathy and lung cancer2. Implementation Challenges: Healthcare systems face integration issues with existing workflows and EHR systems3. Regulatory Developments: FDA has approved 18 AI-based diagnostic tools in the past 2 years4. Emerging Trends: Multimodal AI systems combining imaging with patient history show promising resultsWould you like me to elaborate on any of these areas?"
Machine identifies patterns across sources and synthesizes information into coherent themes.
4
Generate Research Report
Create a structured research report with citations.
You: "Please create a comprehensive report with these findings"Machine: "I'll prepare a detailed research report on AI in healthcare diagnostics. Would you prefer:1. A formal academic-style report with full citations2. A business-oriented white paper format3. A presentation with key points and visualizations4. Something else?"
Machine can generate:
Executive summaries
Detailed reports with sections and subsections
Literature reviews with proper citations
Visual presentations of key findings
5
Review and Verify
Verify the accuracy and completeness of the research.
You: "Can you verify the accuracy of the statistics cited in the report?"Machine: "I'll verify all statistics in the report. For example, the claim that 'AI systems show 15-30% improvement in early detection rates' comes from:1. Wong et al. (2023) in JAMA Network Open - reported 16.8% improvement2. Patel & Johnson (2022) in Nature Medicine - reported 28.3% improvement3. Cleveland Clinic (2024) study - reported 22.5% improvementWould you like me to provide direct links to these sources or verify any other specific claims?"
For critical research, Machine can provide links to original sources for manual verification.
This tutorial demonstrates how to connect Machine to external services through APIs.
1
Identify Integration Needs
Determine which external services you need to connect with.
You: "I need to integrate Machine with our company's CRM system"Machine: "I can help you integrate with your CRM. To get started:1. Which CRM system are you using?2. What specific actions do you want to perform? (e.g., create contacts, update deals)3. Do you have API credentials for the CRM?"
Be specific about what data you need to exchange and what actions you want to automate.
2
Obtain API Credentials
Set up and secure the necessary API access.
You: "We use HubSpot CRM and need to create contacts and update deals. I need to get API credentials"Machine: "Great, I can guide you through obtaining HubSpot API credentials:1. Log into your HubSpot account2. Go to Settings > Integrations > API Keys3. Click 'Create key' and provide a description like 'Machine Integration'4. Copy the generated keyOnce you have the key, we'll use it to set up the integration."
Machine provides step-by-step guidance for API setup while keeping security best practices in mind.
3
Configure Integration
Set up the connection between Machine and your external service.
You: "I have the API key now"Machine: "Perfect! To securely configure the HubSpot integration:1. In Machine, go to Settings > Integrations2. Select 'Add New Integration'3. Choose 'HubSpot' from the list4. Enter your API key when prompted5. Select the specific permissions needed (contact creation, deal updates)Would you like me to guide you through testing the connection?"
Machine helps you configure the integration with the appropriate scope and permissions.
4
Create Automation Workflows
Develop workflows that leverage the integration.
You: "I want to automatically create HubSpot contacts when I receive emails from new clients"Machine: "I'll help you set up that workflow. Here's how we'll approach it:1. Create a trigger that activates when you receive emails from unknown addresses2. Extract contact information from the email signature and content3. Check if the contact already exists in HubSpot4. If not, create a new contact with the extracted information5. Send you a notification when a new contact is createdShould we set this up now?"
Machine can design complex workflows that combine multiple services and actions.
5
Test and Refine
Verify that the integration works as expected and make adjustments.
You: "Let's test this with a sample email"Machine: "I'll help you test the integration. Please upload or forward a sample client email, and I'll simulate the workflow to show what information would be extracted and how the HubSpot contact would be created."
Testing helps identify potential issues before deploying the integration in a production environment.