Skip to main content
JS

Jared Santo

4episodes
3podcasts

Featured On 3 Podcasts

All Appearances

4 episodes

AI Summary

→ WHAT IT COVERS Rob Pike's angry response to AI-generated thank-you email sparks discussion about GitHub's monopoly concerns, pricing changes for self-hosted runners, and predictions for 2026 including agent-first web design, vector databases, and AI budget scrutiny. → KEY INSIGHTS - **GitHub Actions Pricing Controversy:** Microsoft announced then walked back fees for self-hosted runners in December 2025, four days after requesting community input. The timing mid-December and charging for externally hosted infrastructure that GitHub doesn't maintain triggered developer backlash, highlighting concerns about GitHub's market dominance and decision-making transparency. - **Agent-First Design Debate:** Tom Tunguz predicts web design will flip to agent-first in 2026, prioritizing AI agent access over human interfaces. The team debates this is premature, noting agents currently excel only at coding and text summarization, not booking flights or hotels, suggesting agent-as-well rather than agent-first remains more realistic. - **Vector Database Architecture:** Grafana Assistant uses vector embeddings for fast semantic search, avoiding context window overflow by having smaller models describe graphs in natural language before feeding to main agent. Tool overloading strategy combines CRUD operations into single manage-dashboards tool with action parameter, reducing surface area and improving agent decision-making. - **AI Cost Threshold Prediction:** Businesses may pay more for AI agents than human labor in 2026, similar to Waymo rides costing thirty-one percent more than Uber yet growing in demand. The premium reflects perceived safety, reliability, and elimination of onboarding, recruiting, training, and management costs associated with human workers. - **Database Access Pattern Stress:** Agents query databases faster and more complexly than humans, creating unprecedented load patterns. Grafana's Loki team faces this challenge as UI and Assistant features effectively DDoS backend systems, requiring innovations in storage formats, indexing strategies, and data architecture specifically optimized for agent consumption patterns. → NOTABLE MOMENT Rob Pike received a Christmas morning AI-generated thank-you email at 5:43 AM from Claude Opus, praising his computing contributions. He responded twelve hours later with profanity-laden rejection, condemning AI companies for environmental impact while ironically thanking him for advocating simpler software, revealing deep frustration with AI industry contradictions. 💼 SPONSORS [{"name": "Fly.io", "url": "https://fly.io"}, {"name": "Namespace", "url": "https://namespace.so"}, {"name": "Notion", "url": "https://notion.com/changelog"}, {"name": "Tiger Data", "url": "https://tigerdata.com"}] 🏷️ GitHub Monopoly, AI Agents, Vector Databases, Agent-First Design, GitHub Actions Pricing, AI Infrastructure Costs

Go Time

Unpop roundup! 2023

Go Time
38 minProducer and Cohost

AI Summary

→ WHAT IT COVERS Go Time reviews 2023's most popular and unpopular opinions from listeners, revealing 62 total submissions with surprising agreement rates and controversial takes on development practices. → KEY INSIGHTS - **Opinion Polling Results:** 37 of 62 opinions were popular, 21 unpopular, showing Go community agrees more than disagrees on technical and non-technical topics submitted throughout year. - **Microservices Architecture:** Start projects with monoliths instead of drawing microservices on whiteboards first - easier to fix design mistakes in single codebase than distributed systems. - **Technical Decision Making:** Fear of looking dumb drives poor architectural choices as engineers stay quiet about crazy decisions to avoid appearing unintelligent in brutal tech industry. - **Tool Selection Strategy:** Best tool for job isn't always best choice - using familiar technology like Postgres instead of specialized tools reduces complexity and maintenance overhead. → NOTABLE MOMENT One developer claimed TikTok provides the most unbiased news available, earning 95% disagreement and becoming the year's most unpopular opinion among Go Time listeners. 💼 SPONSORS [{"name": "Fly.io", "url": "https://fly.io"}, {"name": "Timescale", "url": "https://timescale.com/ai"}, {"name": "JetBrains", "url": "https://jetbrains.com/go"}] 🏷️ Go Programming, Software Architecture, Developer Opinions, Tech Community

Go Time

That's Go Time!

Go Time
88 minFrom ChangeLog

AI Summary

→ WHAT IT COVERS Go Time podcast concludes after 340 episodes spanning six years, with hosts reflecting on community impact, memorable moments, and announcing Fall Through as the spiritual successor. → KEY INSIGHTS - **Podcast longevity strategy:** Multiple rotating cohosts (6-8 regular contributors) enables consistent weekly production by ensuring 2-3 hosts available each recording, preventing burnout and scheduling conflicts that killed earlier formats. - **Community accessibility approach:** Hosting guests with varying experience levels, from beginners to experts, creates inclusive environment where newcomers feel welcomed and veterans share authentic struggles, lowering barriers to participation. - **Technical content balance:** Mix highly technical episodes with social dynamics discussions, non-technical topics like neurodiversity, and silly segments to maintain broad appeal while serving diverse community needs and learning styles. - **Conference integration benefits:** Live game show episodes at conferences create unique content format, strengthen host-audience relationships through face-to-face interactions, and provide networking opportunities that extend podcast's community-building impact beyond audio. - **Authentic hosting philosophy:** Avoid scripted content, embrace natural conversation flow including mistakes and tangents, and maintain genuine friendships between hosts to create trustworthy environment where guests and listeners feel comfortable. → NOTABLE MOMENT Host Matt Ryer admits to arguing positions he doesn't believe in during episodes to provide balanced perspectives, sometimes realizing mid-discussion that he's completely wrong about his stance. 💼 SPONSORS [{"name": "Fly.io", "url": "https://fly.io"}, {"name": "JetBrains", "url": "https://jetbrains.com"}, {"name": "Timescale", "url": "https://timescale.com"}] 🏷️ Go Programming, Podcast Production, Developer Community, Tech Broadcasting, Software Engineering

AI Summary

→ WHAT IT COVERS Chris Benson defines true autonomous swarming technology, distinguishing it from simple drone fleets, while exploring physical AI applications in home automation, robotics, and the technical challenges of creating distributed decision-making systems. → KEY INSIGHTS - **Swarming Definition:** True swarming requires numerous independent, fully autonomous platforms exhibiting coordinated emergent behaviors as a single distributed decisioning entity - most current "swarm" applications are actually just coordinated fleets with predetermined paths. - **Open Models Advantage:** The performance gap between frontier models from major companies and open source alternatives has narrowed dramatically, making AI model creation increasingly commoditized while pushing companies toward specialized vertical services. - **Physical AI Revolution:** Small, purpose-built robots costing under $30 will proliferate in homes for specific tasks like cleaning, security, and maintenance, communicating via Matter protocol for local, privacy-focused automation without cloud dependencies. - **Development Stack:** Start swarming projects with ROS 2 (Robot Operating System), Rust programming language for embedded systems, Embassy runtime for hardware without operating systems, and small models from Hugging Face for local AI inference. - **Human Control Levels:** "Human on the loop" supervision allows swarms to operate autonomously within mission parameters, while "human in the loop" requires explicit approval for each action - the former enables true emergent swarm behavior. → NOTABLE MOMENT Benson reveals his prediction methodology by admitting his spectacularly wrong 2019 assessment that investors had missed NVIDIA's opportunity, leading a colleague to contrarily invest and profit significantly from the subsequent AI boom. 💼 SPONSORS [{"name": "Miro", "url": "miro.com"}, {"name": "Shopify", "url": "shopify.com/practicalai"}] 🏷️ Autonomous Swarming, Physical AI, Home Automation, Robotics, Rust Programming, Distributed Systems

Explore More

Never miss Jared Santo's insights

Subscribe to get AI-powered summaries of Jared Santo's podcast appearances delivered to your inbox weekly.

Start Free Today

No credit card required • Free tier available