← Back to Showcase

How to Max Out Claude Max

Most people barely scratch the surface of their $200/month Claude Max subscription. Here's the complete playbook for running at 100% usage — from basic settings to running 30 concurrent sessions overnight.

Contents

01 The 4 Usage Levels — Where Are You? 02 Unlock Full Autonomy — Permissions Settings 03 MCPs, Connectors & External Tools 04 claude-fleet — Parallel Task Orchestration 05 Remote Access — Run Tasks While You Sleep 06 Common Inefficiencies & How to Fix Them 07 Pro Tips for Maximum Throughput 08 Case Study — 403 Tasks in 48 Hours
"In evolution, there were animals which were way more efficient without utilizing a big brain.. but mammals, humans even being clumsy in many ways, were able to surpass because of brain and ability to handle complexity. We have new big digital brains now. Whoever utilizes the exponential compute is going to be the winner." — Vimal

The 4 Usage Levels

Most people are stuck at Level 1 or 2. The jump to Level 3/4 is where real value extraction happens.

Level 1 — Chat User

5-15%

Using claude.ai for questions, writing, and code snippets. One conversation at a time. Waiting for responses before asking the next thing.

  • Single chat window
  • Manual copy-paste of code
  • No tool use or file access
  • ~20 messages/hour

Level 2 — Claude Code User

20-40%

Using Claude Code CLI in your project. Asking it to edit files, run tests, search code. Still one session at a time, waiting for each task.

  • Single Claude Code session
  • Approving each tool use manually
  • Small, scoped tasks
  • ~5-10 tasks/hour

Level 3 — Autonomous Sessions

50-80%

Full tool permissions enabled. Large, complex prompts. Multiple parallel sessions using claude --dangerously-skip-permissions or proper allow-lists. Claude works while you review.

  • 3-5 concurrent sessions
  • Auto-approved tool use
  • Multi-file, multi-step tasks
  • ~20-40 tasks/hour

Level 4 — Fleet Orchestration

95-100%

Using claude-fleet to run 10-30 concurrent Claude sessions. Tasks queued overnight. Remote access via OpenClaw. Reviewing outputs, not writing prompts.

  • 10-30 concurrent sessions
  • Automated task pipelines
  • 24/7 execution (overnight)
  • ~100-400 tasks/day

Unlock Full Autonomy

The single biggest bottleneck for most people: Claude stops and asks for permission on every file read, edit, and command. Fix this first.

~/.claude/settings.json
{
  "permissions": {
    "allow": [
      "Bash",
      "Read",
      "Edit",
      "Write",
      "Glob",
      "Grep",
      "Web Fetch",
      "Web Search"
    ]
  }
}
What This Does

Pre-approves all core tools so Claude can read files, edit code, run shell commands, search the web, and navigate your codebase without stopping to ask. This alone can 3-5x your throughput because Claude chains actions without waiting.

For fully autonomous headless sessions (Level 4)
claude --dangerously-skip-permissions -p "your task prompt here"
Per-Project Permissions

You can also set permissions per-project in .claude/settings.json within your repo. This lets you allow npm test and npm run build without giving blanket Bash access everywhere. Add specific commands to the allow list: "Bash(npm test)", "Bash(npm run build)"

MCPs, Connectors & External Tools

The more context and tools Claude can access, the bigger the tasks it can handle autonomously. These connectors turn Claude from a code assistant into a full-stack operator.

GH
GitHub MCP

Search issues, create PRs, read repo contents, manage branches. Essential for autonomous PR workflows.

FS
Filesystem MCP

Read/write files across multiple repos. Critical for cross-repo audits and large-scale refactors.

WB
Web Fetch / Web Search

Research APIs, documentation, competitors. Claude can read docs and apply them in context without you copy-pasting.

SL
Slack MCP

Post updates, read channels, notify on task completion. Useful for fleet status updates.

DB
Database MCPs (Postgres, ClickHouse)

Query production data directly for analysis. Used in NammaYatri audit to analyze 50B+ ClickHouse rows.

LN
Linear / Jira MCP

Read tickets, create issues, link PRs to tasks. Makes Claude aware of project management context.

SH
Bash (Shell Access)

Run tests, build, deploy, git operations, package management. The most important tool for autonomy — Claude can verify its own work.

GF
Grafana / Monitoring MCP

Check dashboards, query metrics. Used for observability analysis in platform audits.

Example MCP config (~/.claude/settings.json)
{
  "mcpServers": {
    "github": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-github"],
      "env": { "GITHUB_TOKEN": "ghp_..." }
    },
    "filesystem": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-filesystem", "/path/to/repos"]
    },
    "postgres": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-postgres", "postgresql://..."]
    }
  }
}
The Connector Multiplier Effect

Each connector you add expands Claude's autonomy radius. Without GitHub MCP, Claude writes code but you manually create PRs. With it, Claude goes from code → commit → PR → done. Without database access, Claude guesses at data patterns. With it, Claude queries real data, finds actual bugs, and writes targeted fixes.

claude-fleet — The 10x Multiplier

The secret weapon for hitting 100% usage. Run 10-30 concurrent Claude Code sessions with a single command.

Write task files
fleet-convert
Run 10-30 sessions
TUI dashboard
Review & merge

What It Is

A parallel task orchestrator for Claude Code. You write task descriptions in markdown files, and claude-fleet converts them into concurrent Claude Code sessions, each working independently on its own task.

View claude-fleet Presentation →

How It Works

  • Task files: Write what you want done in .md files
  • fleet-convert: Converts tasks to Claude Code prompts
  • Parallel execution: 10-30 sessions run simultaneously
  • TUI dashboard: Real-time status of all sessions
  • Tool control: Configure which tools each session can use

Why It Maxes Usage

Claude Max throttles per-session, not per-account (in practice). Running 10 sessions means 10x the token throughput. Running 30 sessions overnight means you wake up to 200+ completed tasks.

This is how Mags ran 403 tasks across 15 repos in 48 hours.

Getting Started

  • Clone: github.com/nammayatri/claude-fleet
  • Write task files in a directory
  • Run fleet-convert && fleet-run
  • Monitor via TUI dashboard
  • Review outputs when done
Example: Run an audit across 5 repos
# 1. Write task files
cat > tasks/audit-backend.md << 'EOF'
Audit the backend codebase for:
- Security vulnerabilities (OWASP top 10)
- Performance bottlenecks (N+1 queries, missing indices)
- Error handling gaps (unhandled exceptions, missing retries)
- Code quality (dead code, TODOs, deprecated APIs)

Output a structured report with severity, location, and fix plan.
EOF

# 2. Create tasks for each repo
for repo in backend frontend mobile infra shared; do
  cp tasks/audit-backend.md "tasks/audit-${repo}.md"
done

# 3. Convert and run
fleet-convert --tasks-dir tasks/ --output fleet-tasks/
fleet-run --concurrency 10 --tasks-dir fleet-tasks/

Remote Access — Run Tasks While You Sleep

The final piece: a machine that runs 24/7 so you can queue tasks at night and review results in the morning.

Hardware

  • Mac Mini M4 Pro — always-on, low power
  • Enough RAM for 30 concurrent Claude Code processes
  • Fast SSD for git operations
  • Stable internet connection

Remote Access

  • OpenClaw — remote terminal access from any device
  • SSH into the machine from your phone/iPad
  • Start fleet tasks, check progress, review outputs
  • Screen/tmux sessions persist across disconnects

The Overnight Workflow

10 PM: Queue 50-100 tasks via fleet-convert
10:05 PM: Start fleet-run with concurrency 20
Sleep.
7 AM: SSH in, check TUI dashboard
7:30 AM: Review outputs, raise PRs for the good ones

Happy To Remotely Manage

Want help setting up your own fleet? Mags is happy to remotely manage your Mac Mini or cloud VM to get claude-fleet running. SSH in, configure the environment, run a test batch, and hand it off to you.

Message on WhatsApp →

Common Inefficiencies & How to Fix Them

Where most people leak usage capacity without realizing it.

1. Waiting for Approval on Every Action

If you're clicking "Yes" on every file read and edit, you're spending 60% of your time watching a spinner. Fix: Set up the permissions allow-list above. Pre-approve Bash, Read, Edit, Write, Glob, Grep.

2. One Session at a Time

Claude Max supports multiple concurrent sessions. Running one session is like having 30 CPUs and using one. Fix: Open 3-5 terminal tabs, each with its own Claude Code session on different tasks. Or use claude-fleet.

3. Small, Micro-Scoped Tasks

"Fix this typo", "add a semicolon", "rename this variable" — these burn a full message round-trip on work that takes 2 seconds manually. Fix: Batch small tasks into bigger prompts. "Audit this file for all typos, naming issues, and style violations. Fix them all."

4. Not Using CLAUDE.md Files

Without a CLAUDE.md, Claude re-discovers your project structure, conventions, and preferences every session. That's wasted tokens on context-building. Fix: Write a CLAUDE.md in your repo root with architecture, commands, conventions, and key files. Claude reads it automatically.

5. Idle Time Between Tasks

Finishing one task, thinking about the next, typing it out. Every gap is wasted subscription time. Fix: Pre-write task descriptions in a file. When one finishes, paste the next immediately. Or use fleet for zero-gap execution.

6. Not Running Overnight

Your subscription runs 24/7 but you only use it 8 hours. That's 67% waste. Fix: Set up a Mac Mini / cloud VM. Queue tasks before bed. Review in the morning.

7. Repeating Context in Every Conversation

Explaining your project, tech stack, and goals every time you start a new session. Fix: Use CLAUDE.md + auto-memory. Claude remembers across sessions. Also use /compact strategically to extend long conversations.

8. Not Using the Right Scale of Projects

Simple projects with few files don't generate enough tool calls to burn through usage. Fix: Target large codebases (10K+ lines), multi-repo work, platform audits, or cross-cutting refactors. Claude shines on tasks with high complexity and many files.

Pro Tips for Maximum Throughput

Techniques that compound over time.

Prompt Engineering for Autonomy

  • Give Claude the full context upfront — don't drip-feed
  • End prompts with a clear deliverable: "Output a file at X"
  • Use "Do not ask me questions, make your best judgment" for headless sessions
  • Include success criteria: "The tests must pass"

Task Design for Parallelism

  • Break work into independent, non-conflicting tasks
  • Each task should work on different files/dirs
  • Use separate git branches per fleet task
  • Merge sequentially after review

Context Window Management

  • Use /compact to compress context mid-conversation
  • Start new sessions for unrelated tasks (don't overload one)
  • CLAUDE.md auto-loads context without wasting messages
  • Use --print flag for one-shot tasks (no conversation overhead)

Review Workflow

  • Let Claude generate, you review — not the other way around
  • Use git diff to review, not reading Claude's explanation
  • Reject and re-prompt is faster than guiding step-by-step
  • Trust Claude on boilerplate, scrutinize on business logic

Case Study: 403 Tasks, 48 Hours, 15 Repos

How one person ran a full-platform audit of NammaYatri — India's largest open-source ride-hailing platform.

MetricBefore Claude MaxWith Claude Max + Fleet
PRs per day0.412+ (30x)
Repos covered1-215
Reports generated~5/week268 total
Concurrent sessions110-30
Hours active per day8-1024 (overnight fleet)
Scope of workSingle feature/bugfixPlatform-wide audit + fixes
The Key Insight

It's not about typing faster or being a better programmer. It's about task decomposition + parallel execution + 24/7 runtime. Break the work into 400 independent tasks, let 30 Claude instances chew through them overnight, and review in the morning. The bottleneck shifts from "writing code" to "reviewing code" — and review is faster.

See Full Showcase → View Platform Audit →

Ready to Max Out?

Start with Step 1: set up your permissions. Then try 3 concurrent sessions. Then try fleet.

claude-fleet Presentation → Get Setup Help →