Author: Will Tygart

  • Claude Context Window Explained: From 200K to 1M Tokens

    Claude Context Window Explained: From 200K to 1M Tokens

    Last refreshed: May 15, 2026

    Model Accuracy Note — Updated May 2026

    Current flagship: Claude Opus 4.7 (claude-opus-4-7). Current models: Opus 4.7 · Sonnet 4.6 · Haiku 4.5. Claude Opus 4.7 (claude-opus-4-7) is the current flagship as of April 16, 2026. Where this article references Opus 4.6 or earlier models, those references are historical. See current model tracker →. See current model tracker →

    Claude AI · Fitted Claude
    Updated April 2026: Claude Sonnet 4.6 and Opus 4.6 now support a 1 million token context window at standard pricing. Haiku 4.5 supports 200,000 tokens. The information below has been updated to reflect current specs.

    Claude’s context window is one of its most practically important technical specifications — and one of the least well understood. This guide explains tokens and context windows, how Claude’s compare to competitors, and strategies for working effectively within context limits.

    What Is a Context Window?

    A context window is the total amount of text a model can process in a single session — everything it can “see” and reason about at once. Context is measured in tokens. As a practical rule: 1,000 tokens ≈ 750 words.

    Claude’s Context Windows

    Access Method Context Window Approx. Words
    Standard Claude (all plans) 1,000,000 tokens (Sonnet/Opus), 200,000 (Haiku) ~750,000 words (Sonnet/Opus)
    Enterprise Claude 500,000 tokens ~375,000 words
    Claude Code 1,000,000 tokens ~750,000 words

    What Fits in 200K Tokens?

    • A full-length novel (~100,000 words)
    • 100-200 typical business emails
    • 10-15 long research papers
    • An entire small codebase (5,000-10,000 lines)
    • A year’s worth of meeting notes from a small team

    PDF and Document Token Costs

    • PDFs: 1,500-3,000 tokens per page
    • Plain text: ~1 token per 4 characters
    • Images: 1,000-4,000 tokens per image
    • Code files: 500-2,000 tokens per file

    Strategies for Long Contexts

    • Extract before uploading: Only upload relevant PDF sections, not full documents
    • Use Projects for reference material: Store knowledge base docs in Projects rather than re-uploading every session
    • Auto compaction (Claude Code beta): When coding sessions approach limits, Claude automatically summarizes history to continue

    Frequently Asked Questions

    How many pages can Claude read at once?

    With 200K tokens and ~1,500-3,000 tokens per PDF page, roughly 65-130 pages while leaving room for conversation.

    Does Claude forget things in long conversations?

    Not within the context window. In very long conversations approaching the limit, older content may be truncated.


    Need this set up for your team?
    Talk to Will →

  • Anthropic IPO 2026: What’s Confirmed, What’s Rumored, and Where to Track It

    Anthropic IPO 2026: What’s Confirmed, What’s Rumored, and Where to Track It

    Last refreshed: May 15, 2026

    ⚠️ No confirmed IPO date exists as of May 8, 2026. Anthropic has not filed an S-1, set a ticker, or announced a listing date. What exists are credible reports of a Q4 2026 target — but no official confirmation. Everything below is sourced and dated. Click any link to get the latest.

    Where Things Actually Stand

    Anthropic is widely expected to pursue an IPO, and the signals are real — but no date has been set. Here is what is confirmed versus what is reported:

    Confirmed Facts (Primary Sources)

    • Current valuation: $380 billion — set in the February 2026 Series G round led by GIC and Coatue. This is the last confirmed, announced valuation. (CNBC, April 29 2026)
    • Revenue run rate: $30B+ annualized — confirmed by Anthropic directly in May 2026. Sources with knowledge of financials put the real figure closer to $40B. (TechCrunch, April 29 2026)
    • IPO law firm engaged: Wilson Sonsini hired to prepare for a potential public listing — confirmed by the Financial Times in December 2025.
    • Preliminary bank conversations: Anthropic has held early-stage talks with investment banks — confirmed by multiple sources, no banks named publicly.
    • No S-1 filed. The SEC has received no public filing from Anthropic as of this writing.

    Reported But Unconfirmed

    • Q4 2026 IPO target — discussed by Anthropic executives internally according to The Information. Bankers reportedly expect the offering could raise more than $60 billion. (TECHi, sourcing The Information)
    • ~$900 billion valuation round in progress — as of April 30, 2026, TechCrunch reported Anthropic was asking investors to submit allocations within 48 hours for a ~$50 billion raise at a $850–$900 billion valuation. A board decision was expected in May 2026. Anthropic declined to comment. (TechCrunch, April 30 2026)
    • October 2026 — cited in some reports as the earliest possible listing window. Not confirmed by Anthropic.
    • $60B+ raise — reported figure for the eventual IPO offering size. Unconfirmed.

    The Valuation Trajectory

    The speed of Anthropic’s private-market repricing is unlike anything in recent tech history:

    • March 2025: $61.5 billion (Series D, led by Lightspeed)
    • September 2025: $183 billion (Series F)
    • February 2026: $380 billion (Series G, led by GIC and Coatue)
    • May 2026: ~$900 billion reportedly under discussion — not yet closed

    Some early backers are reportedly skipping the current round specifically to wait for IPO pricing — a signal that sophisticated money sees the public listing as potentially more attractive than another late-stage private markup.

    Why There’s No Confirmed Date Yet

    Anthropic is a public benefit corporation, which adds governance complexity to any listing. The company is also in the middle of closing what may be its final private round — and closing a $50 billion raise takes time. Until an S-1 is filed with the SEC, no IPO date is official. PitchBook analyst Kyle Stanford has noted that a crowded private financing cycle could push a listing into 2027 if the current round takes longer than expected.

    Who Owns Anthropic Before Any IPO

    Major confirmed investors include Amazon (up to $50 billion committed), Google (up to $40 billion committed), Nvidia ($30 billion), SoftBank ($30 billion), plus Accel, BlackRock-affiliated funds, Fidelity, General Catalyst, Goldman Sachs Alternatives, JPMorganChase, Lightspeed, Menlo Ventures, Morgan Stanley Investment Management, Sequoia, and Temasek. More than 1,000 enterprise customers now spend over $1 million annually on Claude — a figure Anthropic disclosed publicly in May 2026.

    Keep Up With This Story

    This is a fast-moving situation. The sources below are updated in real time — bookmark them if you want the latest as it breaks:

    Want the deeper picture on who is building this company? Read our analysis of Anthropic’s founders and leadership — the most-read piece on this site in this category.

  • Claude AI Alternatives: 10 Tools for When Claude Isn’t Enough

    Claude AI Alternatives: 10 Tools for When Claude Isn’t Enough

    Last refreshed: May 15, 2026

    Claude AI · Fitted Claude

    Claude is one of the best AI assistants available — but it’s not the right tool for every job. It can’t generate images, doesn’t have default real-time web access, and lacks deep Google Workspace integration. Here are the 10 best Claude alternatives, each matched to where it genuinely wins.

    1. ChatGPT — Best All-Around Alternative

    Use when: You need image generation (DALL-E), broader plugin ecosystem, or voice mode. Price: Free / $20/month Plus / $200/month Pro.

    2. Perplexity — Best for Real-Time Research

    Use when: You need current information with source citations. Searches the live web in real time. Price: Free / $20/month Pro.

    3. Gemini — Best for Google Workspace

    Use when: You live in Gmail, Docs, Sheets, or Drive. Native integration across all Google Workspace apps. Price: Free / $20/month Advanced.

    4. Midjourney — Best for AI Image Generation

    Use when: You need high-quality AI-generated images. Claude cannot generate images at all. Price: $10-120/month.

    5. GitHub Copilot — Best IDE-Native Coding

    Use when: You want AI coding assistance embedded in VS Code or JetBrains with persistent autocomplete. Price: $10/month individual.

    6. Otter.ai — Best for Audio Transcription

    Use when: You need to transcribe meetings or audio files. Claude cannot process audio directly. Price: Free / from $10/month.

    7. Jasper — Best for Marketing Content at Volume

    Use when: You’re a marketing team producing high volumes of structured content with brand voice memory and SurferSEO integration. Price: From $49/month.

    8. Microsoft Copilot — Best for Office 365

    Use when: Your work lives in Word, Excel, PowerPoint, Teams, and Outlook. Native M365 suite integration. Price: $30/user/month.

    9. Notion AI — Best for Workspace-Embedded Writing

    Use when: You want AI assistance directly inside Notion — summarizing pages, drafting within documents, auto-filling databases. Price: $8-10/month add-on.

    10. DeepSeek — Best for Cost-Sensitive API Use

    Use when: Building API applications where per-token cost is the primary constraint and you’re not handling sensitive data. DeepSeek API is 10-20x cheaper. Note data sovereignty considerations. Price: Free consumer / very cheap API.

    Frequently Asked Questions

    What is the best free alternative to Claude AI?

    Gemini has the most generous free tier with capable model access. Perplexity free includes limited Pro searches. ChatGPT free uses GPT-4o-mini.


    Need this set up for your team?
    Talk to Will →

  • Claude Max Plan: Who Actually Needs $100/Month

    Claude Max Plan: Who Actually Needs $100/Month

    Last refreshed: May 15, 2026

    Claude AI · Fitted Claude

    The jump from Claude Pro to Max is a 5x price increase — $20/month to $100/month. Whether it’s worth it depends entirely on how you use Claude and where your current plan fails you. Here’s the data to make that decision.

    What’s Actually Different

    Feature Pro ($20/mo) Max 5x ($100/mo) Max 20x ($200/mo)
    Usage volume Baseline 5x Pro 20x Pro
    Heavy prompts/day ~12 ~60 ~240
    Claude Code No Yes Yes
    Extended thinking Limited Full Full
    Model access Sonnet + Opus Sonnet + Opus Sonnet + Opus

    Key insight: you don’t get different models at Max — you get more of them. The difference is usage capacity and Claude Code access.

    Who Should Stay on Pro

    • You use Claude regularly but not all day — a few substantive sessions per week
    • You’re hitting limits occasionally but not consistently
    • You don’t need Claude Code

    Who Needs Max 5x

    • You hit Pro limits daily and it disrupts your workflow
    • You want Claude Code — only available at Max tiers
    • Claude is your primary work tool, not supplementary

    Who Needs Max 20x

    • Heavy Claude Code user running multi-hour sessions daily
    • Processing massive document volumes — dozens of long PDFs per day
    • You’ve been hitting Max 5x limits regularly

    Frequently Asked Questions

    What does Claude Max include that Pro doesn’t?

    Claude Code access, higher usage limits (5x or 20x), full extended thinking, and higher priority during peak times.

    Is Claude Max worth $100 a month?

    For developers using Claude Code and professionals hitting Pro limits daily: yes. For moderate users: Pro at $20/month is sufficient.


    Need this set up for your team?
    Talk to Will →

  • Claude vs Perplexity: Research Engine vs Reasoning Partner

    Claude vs Perplexity: Research Engine vs Reasoning Partner

    Last refreshed: May 15, 2026

    Claude AI · Fitted Claude

    Comparing Claude to Perplexity is a category error — they’re not trying to do the same thing. Perplexity is a real-time research engine. Claude is a reasoning partner. Understanding the distinction helps you build the most effective research workflow.

    What Perplexity Does Best

    • Real-time information: Searches the live web, summarizes current events with source links
    • Source citation: Every claim has source links for verification
    • Quick research: Fast sourced answers for “what is X” and “what happened with Y”
    • Academic research: Academic mode searches peer-reviewed papers

    What Claude Does Best

    • Deep reasoning: Complex multi-step analysis and strategic thinking
    • Document synthesis: Upload a 200-page report and ask for analysis — Perplexity cannot do this
    • Writing quality: Significantly stronger long-form writing
    • Code: One of the best coding models. Perplexity is not a coding tool.
    • Private documents: Works with confidential content you upload

    The Hybrid Workflow (Best of Both)

    1. Perplexity first: Rapid research, current information, source discovery
    2. Claude second: Synthesis, analysis, writing. Take what Perplexity found and reason through the implications

    At $20/month each, running both costs $40/month — worth it for professionals who research and write regularly.

    Frequently Asked Questions

    Should I use Claude or Perplexity for research?

    Use Perplexity for finding current information with sources. Use Claude for analyzing, synthesizing, and writing. Ideally, use both — Perplexity first, Claude second.

    Does Claude have real-time web access?

    Not by default. Claude has a knowledge cutoff and doesn’t browse the web in real time unless connected via MCP or specific integrations.


    Need this set up for your team?
    Talk to Will →

  • Claude vs DeepSeek: Performance, Pricing, and Privacy

    Claude vs DeepSeek: Performance, Pricing, and Privacy

    Last refreshed: May 15, 2026

    Model Accuracy Note — Updated May 2026

    Current flagship: Claude Opus 4.7 (claude-opus-4-7). Current models: Opus 4.7 · Sonnet 4.6 · Haiku 4.5. Claude Opus 4.7 (claude-opus-4-7) is the current flagship as of April 16, 2026. Where this article references Opus 4.6 or earlier models, those references are historical. See current model tracker →. See current model tracker →

    Claude AI · Fitted Claude

    DeepSeek emerged as the most disruptive AI development since GPT-4 — a Chinese lab producing frontier-quality models at dramatically lower cost. In 2026, it’s a genuine competitor to Claude in several categories. But the comparison isn’t only about performance. Privacy and data sovereignty matter. This guide covers all three dimensions.

    Performance Comparison

    Benchmark Claude Opus 4.6 DeepSeek
    SWE-bench (coding) 80.8% ~49% (V3), higher for R1
    GPQA Diamond 91.3% Competitive
    Math reasoning Top tier R1 leads on pure math
    Context window 200K tokens 128K tokens

    Claude leads on real-world software engineering and long-document reasoning. DeepSeek R1 is competitive or superior on pure math. For most professional use cases, Claude holds the performance edge.

    Pricing Comparison

    DeepSeek’s API pricing is 10-20x cheaper than Claude’s — roughly $0.27-0.55 per million input tokens vs Claude’s $3-15. For high-volume API applications where cost is the primary constraint, DeepSeek is a serious consideration. The consumer interface is free vs Claude’s $20-200/month paid tiers.

    The Privacy Question

    DeepSeek is a Chinese company. Its data handling is subject to Chinese law, which includes requirements to provide user data to Chinese government authorities under certain circumstances. Multiple national governments have restricted DeepSeek on government systems. For professionals handling confidential client data or sensitive business information, the data sovereignty difference between Anthropic (US-incorporated) and DeepSeek (Chinese-incorporated) is material.

    Choose Claude If You…

    • Handle confidential professional, legal, or medical data
    • Need highest performance on software engineering tasks
    • Require long-document analysis (200K vs 128K context)
    • Need US-based data handling

    Frequently Asked Questions

    Is DeepSeek as good as Claude?

    Competitive on math and logic. Claude leads on SWE-bench software engineering, long documents, and writing quality.

    Is DeepSeek safe to use?

    For general consumer use, immediate risk is low. Professionals handling sensitive data should consider DeepSeek’s Chinese data jurisdiction carefully.


    Need this set up for your team?
    Talk to Will →

  • MCP Servers Explained: Model Context Protocol Tutorial

    MCP Servers Explained: Model Context Protocol Tutorial

    Last refreshed: May 15, 2026

    Claude AI · Fitted Claude

    Model Context Protocol (MCP) is the most important infrastructure development in Claude’s ecosystem in 2026. It’s an open standard for connecting AI models to external tools, data sources, and services — replacing fragmented one-off integrations with a universal interface. This guide explains what MCP is and how to set up your first server.

    What Is MCP?

    MCP defines a universal interface: any tool that implements the MCP server specification can connect to any AI application implementing the MCP client specification. Build once, connect anywhere. Before MCP, connecting Claude to external systems required custom integration code for every integration — and none of it worked across different AI tools.

    MCP Architecture

    • MCP Host: The AI application (Claude desktop, Claude Code, your custom app)
    • MCP Client: Built into the host; manages connections to servers
    • MCP Server: Lightweight program exposing tools, resources, or prompts

    Setting Up MCP in Claude Desktop

    Go to Settings → Developer → Edit Config. Add your server configuration:

    {
      "mcpServers": {
        "filesystem": {
          "command": "npx",
          "args": ["-y", "@modelcontextprotocol/server-filesystem", "/path/to/directory"]
        }
      }
    }

    Restart Claude Desktop. Claude can now read, write, and manage files in your specified directory.

    Popular MCP Servers

    Server What It Does
    Filesystem Read/write local files
    GitHub Manage repos, issues, PRs
    PostgreSQL Query databases
    Slack Read/send messages
    Brave Search Real-time web search
    Zapier Connect to 8,000+ apps

    Frequently Asked Questions

    Is MCP open source?

    Yes. Anthropic open-sourced the MCP specification and official server implementations.

    Do I need to code to use MCP?

    To install existing servers: basic command-line comfort is enough. To build custom servers: TypeScript or Python knowledge required.


    Need this set up for your team?
    Talk to Will →

  • Claude API Tutorial: Python and JavaScript Getting Started

    Claude API Tutorial: Python and JavaScript Getting Started

    Last refreshed: May 15, 2026

    Model Accuracy Note — Updated May 2026

    Current flagship: Claude Opus 4.7 (claude-opus-4-7). Current models: Opus 4.7 · Sonnet 4.6 · Haiku 4.5. Claude Opus 4.7 (claude-opus-4-7) is the current flagship as of April 16, 2026. Where this article references Opus 4.6 or earlier models, those references are historical. See current model tracker →. See current model tracker →

    Claude AI · Fitted Claude

    The Claude API gives you programmatic access to Claude in your own applications and scripts. This guide gets you from zero to a working integration in Python or JavaScript.

    Prerequisites

    • Anthropic account at console.anthropic.com
    • API key from Console → API Keys
    • Python 3.7+ or Node.js 18+

    Installation

    # Python
    pip install anthropic
    
    # JavaScript
    npm install @anthropic-ai/sdk

    Your First API Call (Python)

    import anthropic
    
    client = anthropic.Anthropic(api_key="your-api-key-here")
    
    message = client.messages.create(
        model="claude-sonnet-4-6",
        max_tokens=1024,
        messages=[{"role": "user", "content": "Explain APIs in plain English."}]
    )
    print(message.content[0].text)

    Adding a System Prompt

    message = client.messages.create(
        model="claude-sonnet-4-6",
        max_tokens=1024,
        system="You are a helpful customer support agent for Acme Corp.",
        messages=[{"role": "user", "content": "How do I reset my password?"}]
    )

    Streaming Responses

    with client.messages.stream(
        model="claude-sonnet-4-6",
        max_tokens=1024,
        messages=[{"role": "user", "content": "Write a 500-word blog post about AI."}]
    ) as stream:
        for text in stream.text_stream:
            print(text, end="", flush=True)

    Model Selection

    Model String Best For
    Claude Opus 4.6 claude-opus-4-7 Complex reasoning, coding
    Claude Sonnet 4.6 claude-sonnet-4-6 Balanced everyday tasks
    Claude Haiku 4.5 claude-haiku-4-5-20251001 Fast lightweight tasks

    Frequently Asked Questions

    How much does the Claude API cost?

    Pricing is per token (input and output separately). Check anthropic.com/pricing. Haiku is cheapest, Sonnet offers the best cost/quality balance for most applications.

    Do I need a Claude subscription to use the API?

    No. API access is separate. Create an Anthropic Console account and pay per token used.


    Need this set up for your team?
    Talk to Will →

  • How to Install a Crawl Space Vapor Barrier: Step-by-Step DIY Guide

    How to Install a Crawl Space Vapor Barrier: Step-by-Step DIY Guide

    The Distillery
    — Brew № 2 · Crawl Space

    Installing a crawl space vapor barrier is the most DIY-accessible component of a full encapsulation system — and the one that saves the most money if done correctly. Material cost for a 1,200 sq ft crawl space is $480–$2,400 depending on barrier quality; professional labor for barrier installation alone is $1,000–$2,500. The $1,000–$2,500 in potential savings is real, but only if the installation is done correctly. Improperly installed barriers — unsealed seams, missed penetrations, inadequate wall coverage — provide significantly less protection than a properly installed system. This guide covers the complete installation process step by step.

    Materials and Tools Needed

    Materials

    • Vapor barrier: Minimum 12-mil reinforced polyethylene (for a full encapsulation; 6-mil is insufficient for most real-world crawl spaces). Calculate quantity: crawl space square footage × 1.35 to account for wall coverage and seam overlaps. For a 1,200 sq ft crawl space: 1,200 × 1.35 = 1,620 sq ft of barrier material needed.
    • Seam tape: Compatible reinforced polyethylene tape designed for vapor barrier seaming — not duct tape, not standard packing tape. Must be labeled as compatible with the barrier material. Budget: 4–6 rolls of 3″ × 180′ tape for a 1,200 sq ft crawl space.
    • Mechanical fasteners: Hammer-drive concrete anchors or Hilti pins (powder-actuated) for fastening the barrier to the foundation wall at the top edge. Alternatively, a construction adhesive compatible with polyethylene.
    • Wall termination strip: A plastic or aluminum channel that holds the top edge of the barrier against the wall and provides a clean termination line. Optional but provides a more professional finished appearance.
    • Pipe penetration seals or tape: Pre-cut penetration seals or compatible tape for sealing around pipes, conduit, and columns.
    • Backer rod: For sealing large gaps at the floor-wall joint before applying the barrier.

    Tools

    • Utility knife with extra blades (barrier material dulls blades quickly)
    • Tape measure and chalk line
    • Hammer drill with concrete bit (for mechanical fasteners)
    • Seam roller or J-roller (a wallpaper seam roller) for pressing seam tape firmly
    • Knee pads
    • Bright LED work light
    • N95 respirator, Tyvek coveralls, gloves, and eye protection

    Phase 1: Preparation (Day 1, 2–4 hours)

    Clear the Crawl Space

    Remove everything from the crawl space floor that would create a puncture hazard or prevent full barrier coverage: old vapor barrier material, rocks and concrete rubble, construction debris, and any stored items. Knock down or smooth sharp concrete protrusions from footings and foundation walls. This preparation step is often skipped by quick-service installers but is essential — sharp debris beneath the barrier causes punctures that undermine the entire installation.

    Remove Old Insulation (If Present)

    Deteriorated fiberglass batt insulation between floor joists must be removed before installing a new vapor barrier. Old insulation harbors mold, pest material, and moisture — leaving it above the vapor barrier creates a micro-environment that defeats the moisture control the barrier is intended to achieve. Use heavy-duty contractor bags for removal; expect 4–8 bags for a 1,200 sq ft crawl space. This is unpleasant work but non-negotiable for a quality installation.

    Identify and Plan for All Penetrations

    Walk the crawl space and identify every penetration through the barrier that will be needed: foundation piers, support columns, plumbing pipes, and electrical conduit. Plan the barrier strips to minimize the number of cuts required around each penetration — in many cases, placing the barrier strip to approach a column from one direction allows a simpler cut than if the column is in the middle of a strip.

    Phase 2: Barrier Installation (Day 1–2, 4–8 hours)

    Start at the Back Wall

    Begin at the wall farthest from the access point. This allows the installation to progress toward the exit — you will not be crawling over freshly installed, untaped barrier material as you work. Unroll the first strip from the back wall across the crawl space toward the front.

    Wall Coverage

    The barrier must extend up the foundation wall — not just cover the floor. The minimum wall coverage is 6 inches above the visible soil or moisture line; 12 inches is better practice; the full height of the foundation wall is best practice for a complete encapsulation. At the back wall:

    • Unroll the barrier strip to extend up the back wall to your target height
    • Secure the top edge to the wall using hammer-drive anchors or construction adhesive, spaced every 12–18 inches
    • The barrier lies flat on the ground from the base of the wall toward the access end

    Seam Overlapping and Taping

    Each subsequent strip overlaps the previous strip by a minimum of 12 inches — 18–24 inches is better practice. The overlap seam is the most critical quality point in the installation. Apply seam tape as follows:

    • Ensure both surfaces at the seam are clean and dry before taping — dust and moisture prevent adhesion
    • Apply the tape centered on the overlap, pressing it firmly down the entire length of the seam
    • Use a seam roller or J-roller to apply firm pressure along the entire tape length — hand pressure alone is insufficient for long-term adhesion
    • Check every seam after taping by attempting to lift the tape at multiple points — it should be firmly adhered with no lifting edges

    Sealing Around Penetrations

    Every penetration through the barrier is a potential moisture pathway. For each penetration:

    • Round pipes and conduit: Cut an X or cross in the barrier, pull the flap up around the pipe, and seal with compatible tape wrapped around the pipe and adhered to the barrier surface. Pre-cut penetration seals (rubber pipe collars with adhesive flanges) provide cleaner results for round penetrations.
    • Square columns and piers: Cut the barrier to the perimeter of the pier base. Apply tape along all four sides where the barrier meets the pier surface — press firmly with the seam roller.
    • Odd-shaped penetrations: Use a combination of cuts, patches, and tape to achieve a continuous sealed barrier around the penetration. Take extra time on these — they are the most common point of future moisture intrusion.

    Completing the Side and Front Wall Coverage

    As each strip is laid, the side walls must also be covered. Cut barrier strips to run up the side walls and tape them to the edge of the floor strips. The barrier should cover all ground-contact surfaces — walls included — to create a true continuous envelope. The front wall (nearest the access) is done last, with the barrier running up and being secured at the top edge near the access opening.

    Phase 3: Quality Check Before Closing

    Before the access door is closed, conduct a final walkthrough:

    • Inspect every seam — no lifting tape edges, no gaps in the overlap
    • Inspect every penetration — tape fully adhered on all sides
    • Inspect wall attachment — barrier secured at top, no gaps at floor-wall junction
    • Photograph the completed installation from multiple angles and distances — this creates your baseline documentation for future inspections and any warranty claims

    Frequently Asked Questions

    How long does it take to install a crawl space vapor barrier yourself?

    For a solo homeowner in a standard-height (36″+) crawl space: 2–3 full days for a 1,200 sq ft crawl space, including preparation and cleanup. Low-clearance crawl spaces (under 24″) are significantly slower — add 50–100% to time estimates. Working with one other person reduces time by approximately 30% and significantly reduces the difficulty of handling full barrier rolls in a confined space.

    How do I calculate how much vapor barrier I need?

    Measure the crawl space floor area. Multiply by 1.35 to account for seam overlaps and wall coverage (assuming 12″ of wall coverage on all sides). For a 1,200 sq ft crawl space: 1,200 × 1.35 = 1,620 sq ft of barrier material. Add 10% for waste from cuts around penetrations in complex crawl spaces. Most barrier products are sold in standard roll sizes (e.g., 10′ × 100′ = 1,000 sq ft per roll) — purchase in the next roll increment above your calculated need.

    What is the best tape for sealing crawl space vapor barrier seams?

    Use tape specifically designed and labeled for vapor barrier seaming — typically a reinforced polyethylene tape or a butyl rubber tape compatible with the barrier material. Do not use standard duct tape (it fails in temperature and humidity extremes), packing tape, or general-purpose seam tape. Products from companies like Nashua, Poly-America, and the barrier manufacturers themselves typically offer compatible seam tape. Confirm compatibility on the packaging — some premium barriers require manufacturer-specific tape to maintain the product warranty.


  • Claude Extended Thinking: When and How to Use It

    Claude Extended Thinking: When and How to Use It

    Last refreshed: May 15, 2026

    Claude AI · Fitted Claude

    Extended thinking is Claude’s most powerful reasoning mode — and the one most people never use correctly. This guide explains what extended thinking does, when it genuinely improves outputs, how to enable it, and when you’re better off with a standard prompt.

    What Is Extended Thinking?

    Extended thinking gives Claude a dedicated reasoning phase before generating its final response. Claude works through a problem on “scratch paper” before writing its answer — exploring multiple approaches, identifying errors in its own reasoning, and building a more deliberate chain of thought. In Claude 4.6 models, this is called adaptive extended thinking — Claude dynamically adjusts how much thinking it does based on problem complexity.

    When Extended Thinking Genuinely Helps

    • Complex math and logic problems requiring step-by-step reasoning
    • Multi-step coding tasks with many interdependent components
    • Strategic analysis requiring weighing many variables
    • Difficult research synthesis where accuracy matters most
    • Any task where “think step by step” would help — extended thinking does this automatically

    When Extended Thinking Is Overkill

    • Simple factual questions with clear answers
    • Routine writing tasks (emails, summaries, short copy)
    • Format conversion or data transformation
    • Tasks where speed matters more than depth

    How to Enable Extended Thinking

    In Claude.ai: Look for the thinking toggle before sending your message. Available on Max tiers and higher.

    Via API: Pass "thinking": {"type": "enabled", "budget_tokens": 10000} in your request. Higher budget_tokens allows more thorough reasoning but increases latency and cost.

    What You See During Extended Thinking

    Claude shows a collapsed “thinking” section before its response. Expand it to see the reasoning chain — useful for verifying logic or understanding how Claude approached a problem. The thinking section is exploratory and may contain dead ends; this is normal.

    Frequently Asked Questions

    Does extended thinking always give better answers?

    No. It improves accuracy on complex reasoning tasks but adds latency. For simple tasks, standard mode is faster and just as accurate.


    Need this set up for your team?
    Talk to Will →