15 episodes · 8 Shorts · 7 deep dives · 66 more videos on YouTube
Every episode and video
60-second Shorts for the one idea, deep dives for the whole design. Episodes come with the written spec: architecture, failure mode, fix and trade-offs. The rest of @AI.JoinDev follows, and plays on YouTube.
52 sec
The Runaway Agent
Your agent looped all night. $10k gone.
6 min
Building Memory for AI Agents
Your agent remembered the wrong version of you.
4 min
Autonomous Coding Agents
The hard part is not writing code. It is proving the change is safe.
5 min
Why AI Agents Act Twice
The tool succeeded. Your agent did it again.
45 sec
MCP Tool Poisoning
One bad tool can leak all your keys.
56 sec
Jev vs Laya
Same idea. One closed, one open.
54 sec
Jev: Decisions, Not Text
Stop parsing LLM prose just to get a yes or no.
7 min
Context Engineering for AI Agents
Brilliant at step one. By step forty, your agent forgot the task.
8 min
Jev vs LLM: Decide or Generate
Many LLM calls are decisions dressed up as text.
50 sec
Refresh Token Replay
Someone else is using your AI agent's keys.
45 sec
Post-Filter RAG Leaks
Your RAG bot just leaked salaries.
56 sec
Edge Rendering Trap
Edge rendering made your AI app slower.
55 sec
CDN Cache Stampede
One cache expiry just took down your AI app.
7 min
Why CDNs Make Far Servers Fast
Your server is 13,000 km away. Every hello pays for it.
6 min
Semantic Caching for LLMs
Your AI cache just answered the wrong customer.
0:36LLM Prefix Caching: Stable Prefix First
0:38RPC Deadline Propagation: One Time Budget in 60 Seconds
0:37Cache Stampede Prevention: Request Coalescing in 60 Seconds
0:31Agentic AI Architecture: Build a Bounded Agent Loop
0:36AI Agent Tool Calls: Prevent Duplicate Side Effects
0:31MCP Schema Evolution: Additive First
0:37Request Hedging: Cut Tail Latency Without a Retry Storm
0:33Fencing Tokens: Stop Stale Distributed Lock Holders
0:25Consistent Hashing: Scale Caches Without Remapping Every Key
0:37Queue Autoscaling: Backlog per Worker in 60 Seconds
0:39Transactional Outbox Pattern in 34 Seconds
0:40Duplicate Events: Idempotent Consumer Fix in 60 Seconds
0:45Saga Pattern in 60 Seconds | AI System Design
0:41Priority and Fair Queuing in 60 Seconds
0:42Backpressure in 60 Seconds: Stop Streaming Queue Collapse
0:37Bulkhead Isolation in 60 Seconds: Stop Cascading Resource Failures
0:42Dead-Letter Queue Playbook in 60 Seconds | AI System Design
0:39Retry Storms: Backoff and Jitter in 60 Seconds
0:38Circuit Breaker State Machine in 60 Seconds
0:35OAuth DPoP: Bind Access Tokens to Client Keys in 60 Seconds
0:37Webhook Ingestion: Verify, Persist, Ack Fast
0:30Zero-Downtime Search Reindexing: Atomic Alias Swap
0:37MCP Tasks: Make Long-Running Tool Calls Durable
0:36Out-of-Order Webhooks: Converge State or Sequence Events
0:31ETag + If-Match: Prevent Lost Updates in 60 Seconds
0:35JWT Key Rotation: Use JWKS Overlap to Avoid Auth Outages
0:21Sandboxed Agent Workers: Safer Cloud Coding Agents in 30 Seconds
0:22PostgreSQL SKIP LOCKED Worker Queues in 30 Seconds
0:25Context Eviction & Summary Anchors: In 30 Seconds
0:25Semantic Routing with Confidence Thresholds: In 30 Seconds
0:32Stop AI Video Edits From Drifting
0:31Make AI Transcripts Trustworthy Before Publishing
0:29Turn Team Chat Into a Safe Coding Agent Task
0:39AI Model Migration Ladder: Switch Models Safely #Shorts
0:32Safe Hardware AI Agents: Declare, Check, Act, Watch #Shorts
0:32Token-Efficient AI Agents: Budget, Route, Verify, Stop #Shorts
0:26Spatial Agent Gesture Sync: Make AI Guidance Land #Shorts
0:33Agent Identity Passport: Secure Every AI Action #Shorts
0:28AI Agent Canary Tokens: Catch Unexpected Access Early #Shorts
0:28Human-in-the-Loop AI Agents: Approve What Matters #Shorts
0:25Prompt Contracts: Get Better AI Results Every Time #Shorts
0:27Agent Harness Adapters: Swap Coding Agents Without Rewriting #Shorts
0:36Browser Agent Safety Ladder: Stop Prompt Injection #Shorts
0:26AI Workflow Receipts: Make Automations Reviewable and Replayable #Shorts
0:26LLM Gateway Routing Rules: One Endpoint, Many Models #Shorts
0:27AI Check-for-Understanding Loop: Teach What Comes Next #Shorts
0:26Control AI Video with First and Last Frames #Shorts
0:28Next.js Security Patch Playbook: Upgrade Without Breaking Production #Shorts
0:26Agent Side Chats: Keep Coding Tasks Moving in 30 Seconds
0:36Make Your Website Readable by AI Agents
0:31Next.js 16.3: Stream vs Cache vs Block Navigation
0:30Portable Agent Plugins in 30 Seconds
0:35AI Video Input Ladder: Text, Frames, or References?
0:28Agent Exit Criteria: A Verifiable Definition of Done
0:26Parallel Coding Agents with Git Worktrees
0:26AI Code Review Effort Budgets: Match Depth to Risk
0:25Coding Agent Permission Zones: Sandbox, Approve, Limit
0:25Streaming File Pipelines for AI in 30 Seconds
0:26AI Rubber-Duck Review: Catch What Your First Model Missed
0:27Durable AI Agent State Across Workflow Steps
0:32Optimistic UI Explained in 31 Seconds ⚡ | Faster Web Apps #Shorts
0:24Chunked Prefill for Faster LLM Serving: In 30 Seconds
0:26KV Cache Offloading: More Context per GPU in 30 Seconds
0:29Token Budget Guardrails for AI Agents: In 30 Seconds
0:27MCP Stateless Protocol, Explicit State Handles: In 30 Seconds
0:23Circuit Breakers for AI Tool Calls: In 30 Seconds
Nothing matches yet. Try another word, or clear the filters.
- AI System Design, Visualized on YouTube · 1 videos
- AI System Design Deep Dive on YouTube · 5 videos
- AI System Design in 60 Seconds on YouTube · 30 videos
- AI & Dev Concepts · 30 Seconds on YouTube · 40 videos