Not Replaced Yet

Not Replaced Yet

By Not Replaced YetTechnology
Download on the App Store

Not Replaced Yet episodes

  • AI Agents Ignore Your Skills: Vercel Found the Fix (For Claude Code, Codex, and more)

    Agents are ignoring your AI skill.

    And if your whole workflow depends on that skill firing… that’s a problem.

    In Vercel’s agent evals (testing modern Next.js APIs), the “skill” approach often didn’t get invoked at all so the pass rate didn’t improve. Then they tried a different approach that removes the decision point entirely and it hit a 100% pass rate.

    In this video I break down:

    - Why skills fail to trigger reliably (and why telling the agent to use it is still brittle)
    - The exact pattern that made the agent consistently consult the right docs
    - How they avoided context bloat while keeping perfect accuracy
    - The bigger shift: from pre-training reasoning → retrieval reasoning

    Chapters

    0:00 100% vs 53%
    0:08 Why skills are everywhere right now
    0:34 The eval results that changed my mind
    1:06 The real problem: outdated training data
    1:27 Skills vs “always-on” context (the key difference)
    2:17 Skills didn’t trigger in 56% of cases
    3:04 “Force the skill” helps… but it’s brittle
    3:52 Remove the decision point (the real fix)
    4:21 The results: 100% across build/lint/test
    5:00 How they avoided context bloat
    5:56 Practical recommendations
    6:29 Final thought + question for you

    Watch the video: https://youtu.be/yEg-7sp9GLU

    8 min
  • This is Your Advantage in the AI Era. Win 2026 with Agency.

    In the era of AI, agency is greater than intelligence.

    I kept accidentally landing on the same post in my browser autocomplete. After seeing it over and over, it finally hit: in a world where you can rent genius-level reasoning for the cost of a subscription, intelligence isn’t the scarce resource anymore.

    This video breaks down Andrej Karpathy’s “Agency is greater than Intelligence” idea, what agency actually means, and why this mental model matters right now if you want to stay ahead in the AI era.

    Chapters

    0:00 Agency is greater than intelligence
    0:06 The “muscle memory” internet autopilot
    0:21 The autocomplete mistake that changed everything
    0:53 Too much AI signal to absorb
    1:32 The idea that kept repeating
    1:50 Karpathy + “Agency is greater than Intelligence”
    2:14 What “agency” actually means
    2:33 Why it matters right now
    2:36 Intelligence is being commodified
    2:58 Participant vs spectator
    3:07 Like / comment / subscribe

    Watch the video: https://youtu.be/mkYSbHrnMFU

    4 min
  • Claude Code’s Architecture Is the Feature: How Effective Agents Get Built

    Claude Code’s latest updates aren't about one shiny feature. It’s the same architectural decision showing up in three places: MCP tool search, setup hooks, and hooks that can inject context at the moment of action. The theme is simple: load less at startup, surface more in the moment. This is Progressive Disclosure.

    In this video, we break down how MCP tool search flips the old “load everything upfront” approach (and why that was burning huge chunks of your context window), how setup hooks let initialization run only when it’s actually needed, and why the new hook behavior (returning additional context pre-tool-use) is a big deal for anyone building agentic workflows.

    If you’re wiring up lots of MCP servers, building custom tooling, or designing agent architectures, this update is worth understanding because it’s a pattern you’re going to see everywhere.

    Chapters

    00:00 The “one decision” behind the update
    00:24 Welcome back to Not Replaced Yet
    00:40 Progressive disclosure explained (why it matters)
    01:19 MCP tool search: lazy-loading tools to save context
    02:26 Setup hooks: initialization at the right moment
    02:52 Hooks upgrade: injecting context right before actions
    03:41 The pattern across skills, tools, hooks, and setup
    04:09 Like + subscribe + questions in the comments

    Watch the video: https://youtu.be/ozH00H98w6s

    5 min
  • The 2026 AI Shift: Chatbots → Agents (Claude Code Inside)

    It’s 2026 and your second foundational AI upgrade is to stop using ChatGPT.

    Like a chatbot, that is.

    Chatting back and forth isn’t the unlock anymore. The unlock is getting AI to do things for you, not just tell you things. In this video, I break down the shift from chatbots → agents, show what “agentic” tools can actually do (web browsing, running code, creating files, interacting with apps), and run three real tasks across three agents so you can see the difference in outputs and workflows.

    The big takeaway: stop thinking “What can I ask AI?” and start thinking “What can I delegate to AI?” Build the habit of reaching for agents first.

    Chapters

    00:00 The 2026 AI upgrade: stop “chatting” with AI
    00:43 Chatbots vs Agents (what agents can do that chatbots can’t)
    01:14 A quick tour of agent options (ChatGPT, Manus, Perplexity, Claude Code, more)
    02:32 The underrated skill: context engineering
    02:51 Task #1: competitor research + 1-page report (ChatGPT Agent Mode)
    03:11 Task #2: repurpose a transcript into multi-platform content (Manus)
    03:39 Task #3: investigate a codebase + generate technical docs (Claude Code)
    04:10 Results: what each agent produced (and how it’s delivered)
    07:54 The mindset shift: “What can I delegate?”
    08:20 Wrap-up + what to do next

    Watch the video: https://youtu.be/DryL2DarF4s

    10 min
  • Claude Code v2.1.3: The Mental Model Upgrade (Plus Big Fixes)

    In this episode of Not Replaced Yet, I’m breaking down Claude Code v2.1.3. A surprisingly meaningful release with a real mental-model simplification (slash commands + skills are now one unified concept), plus a bunch of quality-of-life fixes that matter if you’re actually using this thing daily.

    If you’re not using Claude Code yet, you’ll still get the vibe of how fast this tool is evolving. And if you are using it, these are the updates you’ll want to know about.

    Links

    Full Changelog: https://github.com/anthropics/claude-code/blob/main/CHANGELOG.md

    Claude Code: https://code.claude.com/docs/en/quickstart

    v2.1.0 video: https://www.youtube.com/watch?v=njAYrOOMenI

    v2.1.2 video: https://www.youtube.com/watch?v=mFXR-JZUBOI

    Chapters

    00:00 The “/doctor” joke + why it’s useful
    00:14 Intro + why v2.1.3 matters
    00:45 What’s covered in v2.1.3
    01:02 Slash commands + skills merged (mental model simplification)
    01:54 Release channel toggle in /config (stable vs latest)
    02:11 Unreachable permission rules detection (+ fixes via /doctor)
    02:34 VS Code permission destination selector (project/all/team/session)
    03:06 Plan file reset fix after /clear
    03:27 Skill duplicate detection fix (XFAT edge case)
    03:51 Background task count mismatch fix
    04:06 Sub-agent model fixes (correct model during compaction + web search)
    04:33 Trust dialog fix from home directory (hooks properly enable)
    04:57 Terminal rendering stability improvements
    05:08 Slash command description truncation (2-line autocomplete)
    05:29 Hook timeout extended: 60s → 10 minutes
    05:51 Wrap-up + January 2026 pace is wild

    Watch the video: https://youtu.be/nRa77FXf3ow

    7 min
  • Claude Code Is Shipping at an Absurd Pace (v2.1.2 Updates)

    In this episode of Not Replaced Yet, I break down the latest Claude Code updates (v2.1.2). The Anthropic team is shipping at a wild pace.

    If you’re already using Claude Code, these updates remove real friction from everyday dev workflows. And if you aren’t using Claude Code yet, get on board!

    We cover everything from image source metadata and clickable file paths, to massive improvements around large outputs, permissions, Windows installs, and — most importantly — a critical security fix you do not want to skip.

    Chapters

    00:00 Claude Code is shipping fast
    00:50 What’s new in v2.1.2
    01:03 Image source path metadata
    01:35 Clickable file paths in output
    02:01 Shift+Tab auto-accept in Plan Mode
    02:18 Bash output saved to disk (no more truncation)
    02:55 Large tool output persistence
    03:16 Permission system improvements
    03:43 Unified Plugins + MCP view
    04:10 Windows install via Winget
    04:24 ⚠️ Command injection security fix
    05:15 Memory leak fix
    05:34 Binary files loading bug fixed
    05:56 Image paste over SSH improvements
    06:11 macOS Meta key setup guidance
    06:27 Socket file crash fix
    06:43 Final thoughts on Claude Code’s pace

    Watch the video: https://youtu.be/mFXR-JZUBOI

    8 min
  • Claude Code 2.1 Is a Big Upgrade — Here’s What Changed

    In the time it took you to click this video, Claude Code shipped another update 😅

    Today I’m breaking down the major Claude Code 2.1.0 release (and the quick follow-up patch), including the biggest workflow wins: automatic skill hot-reload, forked skill execution, wildcard Bash permissions, hooks everywhere, real-time thinking display, and more.

    If you’re using Claude Code already, these are the changes you’ll feel immediately. If you’re not using it yet… get started!

    Chapters

    00:00 Claude Code shipped again (lol)
    00:13 Welcome to Not Replaced Yet
    00:44 Claude Code 2.1.0 overview (+ quick patch context)
    00:55 Automatic skill hot reload (no more restarts)
    01:14 Forked skill execution (context: fork)
    01:48 Agent type specification in skills
    02:14 Wildcard Bash permissions (* anywhere)
    02:56 Hooks in agent/skill/command front matter
    03:29 Hooks “once: true”
    04:04 MCP list_changed notifications
    04:51 Real-time thinking display (Ctrl/Cmd+O transcript mode)
    05:24 Plugin hook support (global enforcement)
    05:55 /plan shortcut
    06:12 Pre-tool hooks can modify tool inputs (updatedInput)
    06:49 Background task management (Ctrl/Cmd+B)
    07:13 File read token limit override env var
    07:53 Slash command autocomplete anywhere
    08:16 Skills show up in the slash menu
    08:37 Better command search prioritization
    09:10 Reduced permission prompts for chained Bash commands
    09:45 Wrap-up

    Watch the video: https://youtu.be/njAYrOOMenI

    11 min
  • An AI Agent Built My Website… Then Shipped It to Production (Vibing an App into Existence Pt 2)

    60% of the time, vibe coding works every time 😏

    Welcome back to Not Replaced Yet. This is Part 2 of vibing an app into existence. In this episode I jump back into the NRY website build, make quick UI tweaks, add a smooth entrance animation, test the admin/dashboard flow (Clerk auth + Supabase-backed content), and then go from “it works locally” to a real production deploy on Vercel… including hitting (and fixing) a Next.js security/build issue mid-flight.

    Big theme of this series: vibe coding is great when the stakes are LOW!

    Chapters

    0:00 “60% of the time…” + intro
    0:22 Part 2 setup + recap of Part 1
    1:04 The rule of vibe coding: keep stakes LOW
    1:45 /init + cloud.mmd context trick (why it helps)
    2:53 Homepage tweaks (remove background, stack headline, tighten subhead)
    4:54 Adding a soft fade/float-in text animation
    6:39 Fixing animation choppiness (timing + load interaction)
    9:53 Blog + Clerk auth test (finding sign-in route)
    12:50 Creating a blog post via ChatGPT + publishing workflow
    14:54 Testing “Signal” + how links behave
    17:20 Building the Resources page + featured items
    20:23 Creating a new private GitHub repo + first push
    23:15 Importing into Vercel + env vars
    26:56 Vercel build failure (Next.js vulnerability)
    28:49 Security review callout + updating Next.js
    33:10 Why you should be talking to your computer (Whisper Flow)
    35:45 Build succeeds + auto-deploy to production
    37:36 Live site tour (Home / Blog / Signal / Resources)
    38:09 Wrap-up + what’s next

    Watch the video: https://youtu.be/vreNa2rWw_E

    41 min
  • Vibe Coding a Full Website in 2026 (Claude Code + Supabase MCP + Vercel + More!) - Part One

    In this first episode of a two-parter of Not Replaced Yet, I go off-the-cuff with a live vibe code session for the NRY website. We start from zero in Claude Code and let the agent cook.

    Along the way, I show the exact workflow I’m using in 2026:

    • Plan Mode for specs and clarifying questions
    • Design inspiration loops (Dribbble + Landing Love)
    • Copy-paste UI acceleration with 21st.dev prompts
    • Supabase + MCP for database setup and migrations
    • Sub-agents running in parallel with fresh context windows
    • A simple Claude Code hook that plays an ElevenLabs voice when a task finishes
    • Real debugging when things inevitably break (React 19 + hydration issues + styling fixes)

    This is low-stakes build mode — don’t do this for serious client work. But for personal projects and fast iteration, it’s insane.

    Part 2 is coming next, where we finish the remaining UI polish, fix the last issues, and ship the site.

    Chapters

    0:00 Intro and what we’re building
    0:20 What “vibe coding” actually means (and when it’s OK)
    1:24 Why your own website always gets neglected
    2:56 Plan Mode setup and the initial prompt
    5:10 Finding design inspiration on Dribbble + Landing Love
    10:19 Using 21st.dev prompts to speed up UI
    14:08 Supabase setup and adding MCP
    20:39 Context windows and why quality degrades near the limit
    22:36 How the Claude Code creator runs multiple agents in parallel
    29:12 Spinning up sub-agents to do research without burning context
    32:18 Hooks that run when an agent finishes
    34:32 Adding an ElevenLabs audio cue for “task complete”
    45:26 Clerk auth setup (admin-only access)
    54:48 Build completes and what got generated
    55:42 /compact and starting the next phase cleanly
    58:45 Applying the Supabase schema/migration via MCP
    1:00:29 Running the app locally and wiring env vars
    1:04:17 React 19 issue and a full shader rewrite in vanilla Three.js
    1:07:44 First look at the site and quick UI checks
    1:10:57 Fixing the invisible headline (opacity issue) and remaining hydration error
    1:12:16 Wrap-up and Part 2 teaser

    Watch the video: https://youtu.be/jOop4g8_ZZM

    1 hr 14 min
  • Stop Typing: Talking to Your Computer Is the Future

    Your voice is the most important AI upgrade you can make in 2026.

    In this episode of Not Replaced Yet, I break down a new class of AI tools that let you talk to your computer and actually get real work done faster.

    We walk through modern voice-to-text apps like WisprFlow, Superwhisper, and others, why they’re different from basic dictation, and how they quietly change the way you work day-to-day. I also show exactly how I use voice inside my actual workflow, including talking directly into Claude Code.

    ⏱️ Timestamps

    Chapters

    00:00 Why your voice is the #1 AI upgrade in 2026
    00:27 What this new class of voice apps actually does
    00:35 The Tools
    01:43 Pricing, closed vs open source tradeoffs
    02:03 WisprFlow walkthrough
    02:52 Why this isn’t “just dictation”
    03:07 Live examples
    04:37 Final thoughts

    Watch the video: https://youtu.be/EmTHDYB4rs4

    6 min

About Not Replaced Yet

From the publisher's feed

Not Replaced Yet is the show for makers, marketers, and everyday pros in any industry who refuse to let the AI wave roll over them
.
Hosted by brothers Nick (veteran developer) and Jesse Bushkar (growth-obsessed marketer), each episode turns bleeding-edge tech into practical playbooks you can use today. Expect: