Elon doubled limits

2026-05-07 · Source: Ben's Bites · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Robotics & Autonomous Systems, Software Development & Engineering · Depth: Advanced, extended

Summary

Ben Tossell developed a custom email client, replacing Superhuman's $40/month service, primarily for enhanced customization of split inboxes and email rules. His client, built in approximately two hours, uses Gmail's CLI and agent-native features, focusing on label-based archiving and filtering rather than advanced AI for drafting or personal assistance. This build was inspired by Dan Shipper's Codex-native email workflow but tailored to specific needs. Meanwhile, OpenAI released "GPT-5.5 Instant" for free ChatGPT users, offering significant improvements in vision, PDF understanding, web search, and memory, with 52.5% less hallucination. ChatGPT also now integrates with Excel and Google Sheets for analysis and formula generation. Anthropic expanded Claude's usage limits for paid plans by leveraging SpaceX's Colossus 1 data center and introduced new features for Claude Managed Agents, including "Dreaming" for memory, "Outcomes" for agent evaluation, and multi-agent orchestration. Posthog is developing a Codex-like app for self-driving product loops, using product usage data to inform code generation.

Key takeaway

For AI architects evaluating email management solutions, consider building a custom, agent-native client to gain precise control over automation and data handling. This approach allows you to tailor AI assistance to specific tasks like filtering and organization, avoiding unnecessary features and costs associated with generic platforms. Focus on leveraging existing APIs and agent frameworks to create a highly personalized and efficient workflow that aligns exactly with your operational requirements.

Key insights

Custom, agent-native email clients offer superior control over AI-driven features compared to off-the-shelf solutions.

Principles

• Personal software can be built by anyone.
• AI models are evolving rapidly in capabilities.

Method

Building a custom email client involves leveraging existing services like Gmail CLI and agent-native capabilities for rule-based filtering and organization, prioritizing specific user needs over general AI features.

In practice

• Explore custom email solutions for specific workflow needs.
• Utilize new GPT-5.5 Instant for enhanced content understanding.
• Integrate ChatGPT with spreadsheets for data analysis.

Topics

• Custom Email Client
• AI Agents
• 11 Labs
• Voice AI Models
• Generative AI Updates

Best for: CTO, VP of Engineering/Data, AI Architect, AI Engineer, AI Scientist, Director of AI/ML

Related on AIssential

• Ben's Builds #3 - an email app — AI agents can rapidly prototype and refine complex applications, even for non-coders.
• Gmail AI Inbox Looks Very Promising! — AI agents can contextually prioritize emails and generate personalized briefings based on user interactions.
• Making AI Personal: Building Context-Rich AI Tools with MCP • Kris Jenkins • GOTO 2025 — Agentic AI combines LLMs with programmer-defined tools, enabling context-rich answers to ambiguous personal queries.
• Built from the inside out: How AWS Professional Services became a frontier team first — Real productivity gains in AI-native development stem from fundamentally redesigning workflows, not just layering AI onto existing processes.
• Step-by-Step Guide to Building AI Agents Using LLMs — LLM-powered AI agents enable advanced automation by understanding language, reasoning, and interacting with systems autonomously.
• Ad hoc agents are changing how we approach automation — The workflows agent automates agent creation, enabling natural language-driven task automation and calendar integration.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Ben's Bites.