More How I AI episodes

I built a custom Slack inbox. It was easier than youd think. | Yash Tekriwal (Clay) thumbnail

I built a custom Slack inbox. It was easier than youd think. | Yash Tekriwal (Clay)

Published 8 Apr 2026

Duration: 00:44:36

Managing overwhelming Slack notifications through a Kanban-style categorization system, archival tools, and a proposed automated app, while exploring Slack API integrations and structured coding solutions to streamline prioritization and reduce cognitive load.

Episode Description

Yash Tekriwal is the head of education at Clay. A self-described hyper-optimizer, Yash has built multiple custom productivity applications using Perpl...

Overview

The podcast discusses strategies for managing overwhelming Slack notifications, which include over 1000 daily messages, with the majority being non-actionable (60-80%) and only a fraction requiring urgent attention. The author categorizes notifications into three priority levels inspired by a Kanban system: "Red" for urgent tasks (e.g., scheduling requests), "Yellow" for messages needing attention (e.g., updates), and "Green" for passive updates (e.g., social media posts). Slacks "Archive All" feature helps clear non-urgent notifications, while the author envisions a third-party app that automates this process, offering to pay $15/month for such a tool to focus on higher-value work.

Technically, the author developed a custom Slack organization system using Perplexity Computer and Slacks API to filter notifications based on urgency, user-defined criteria (e.g., direct mentions, threads), and timestamps. The discussion emphasizes deterministic solutions over AI-driven approaches, leveraging APIs to categorize messages systematically. Challenges include manually sifting through non-urgent notifications and the inefficiency of Slacks default digest, which the author improved by creating a clean, Kanban-style interface for actionable tasks. The system aims to reduce cognitive load by prioritizing urgent notifications while archiving or ignoring non-critical updates.

The podcast also explores broader productivity trends, such as the rise of niche, low-cost tools to automate repetitive tasks (e.g., email cleanup, meeting note entry) and the potential for modular, customizable workflows using AI. While AI tools like Perplexity Computer are highlighted for their ability to streamline communication and task management, the author emphasizes the importance of deterministic systems and structured APIs for reliable categorization. The content concludes with a vision of outsourcing notification management to dedicated apps, freeing users to focus on collaboration and high-priority work.

Recent Episodes of How I AI

22 Jun 2026 How Claude Mythos found a 15-year-old bug in Mozilla Firefox | Brian Grinstead

Recommended: AI finds bugs

Firefox employs AI agents as "coding archaeologists" to detect and address security vulnerabilities in its massive codebase, leveraging models like Mythos and custom validation tools to identify and systematically fix nearly 500 bugs, while balancing automation with human oversight and open-source collaboration to enhance scalability and security.

15 Jun 2026 How Braintrust uses AI agents, evals, and CI to ship better software | Ankur Goyal

AI integration in software engineering enables agents to handle complex tasks through benchmarking and optimization, shifts engineers toward higher-level work, and addresses challenges like reliability, data parsing, and balancing automation with human expertise while emphasizing outcome-focused systems over procedural methods.

9 Jun 2026 Claude Fable 5 review: what the new Mythos model gets right (and very wrong)

Anthropic's Claude Fable Five excels in long-term technical tasks with strong coding, vision, and async workflow capabilities but faces high token costs, design limitations, and restricted use in cybersecurity/biology, making it suitable for precise, extended projects rather than creative or agile workflows.

More How I AI episodes