Azure Kamsaki oSushi
@kamsakihiyuuma
IT企業で部長をしています。37歳でIT未経験転職から現在4年目、年収は3倍に。クラウド・生成AIシステムの設計・開発と、チームづくり、採用活動をしています。
AWS・Azure・生成AI・ITキャリア・年収・マネジメントについて発信。
中身は美少女野球選手です。
414 Following 769 Followers
AI is getting very good at writing code.
But writing code was never the whole job.
Architecture.
Security.
Observability.
Testing.
Cost.
Operations.
Incident response.
The better AI gets at coding, the more obvious the rest of software engineering becomes.
Show more
I’ve been seeing Supabase more and more in AI application stacks lately.
And it makes sense.
PostgreSQL + pgvector gives you a surprisingly capable foundation for RAG, semantic search, and agent memory — while Supabase handles much of the backend around it.
This 30-day roadmap goes from PostgreSQL basics all the way to production AI SaaS.
Curious: are you using Supabase for AI apps, or do you prefer a separate vector database?
Show more
30 days Supabase Tutorial Series (AI Era Edition)
Master Supabase from beginner to production by building modern full-stack and AI-powered applications every day.
1. Day 1: What is Supabase? – PostgreSQL, Backend-as-a-Service (BaaS), and Why It Matters in the AI Era
Project: Create your first Supabase project.
2. Day 2: Setting Up Supabase – Dashboard, CLI, Local Development, and Project Structure
Project: Set up a local Supabase development environment.
3. Day 3: PostgreSQL Fundamentals – Tables, Rows, Columns, Relationships, and SQL Basics
Project: Build a blog database schema.
4. Day 4: Authentication – Email, Password, Magic Links, OAuth, and Anonymous Login
Project: Build a complete authentication system.
5. Day 5: Row Level Security (RLS) – Policies and Secure Data Access
Project: Secure a multi-user notes application.
6. Day 6: CRUD Operations with Supabase Client Libraries
Project: Build a task management application.
7. Day 7: Database Relationships – Foreign Keys, Joins, and Constraints
Project: Build a school management database.
8. Day 8: Database Functions, Triggers, and Stored Procedures
Project: Automate inventory updates.
9. Day 9: Real-Time Features – Live Database Changes and Presence
Project: Build a real-time chat application.
10. Day 10: Storage – Uploading, Serving, and Securing Files
Project: Build an image gallery.
11. Day 11: Edge Functions – Serverless APIs with Deno
Project: Build a custom REST API.
12. Day 12: Database Migrations and Schema Versioning
Project: Manage database changes using migrations.
13. Day 13: Supabase CLI – Local Development, Linking Projects, and Deployment
Project: Deploy a production-ready backend.
14. Day 14: Full-Text Search and PostgreSQL Extensions
Project: Build a searchable knowledge base.
15. Day 15: Vector Embeddings with pgvector
Project: Build semantic document search.
16. Day 16: Building RAG Applications with Supabase
Project: Create an AI document assistant.
17. Day 17: AI Chat Applications with Supabase and OpenAI
Project: Build an AI chatbot with conversation history.
18. Day 18: AI Agent Memory Using Supabase
Project: Create persistent memory for AI agents.
19. Day 19: Real-Time Dashboards and Live Analytics
Project: Build a live analytics dashboard.
20. Day 20: Supabase with Next.js 15 and React
Project: Build a full-stack SaaS application.
21. Day 21: Supabase with Flutter and React Native
Project: Build a cross-platform mobile app backend.
22. Day 22: Background Jobs, Cron Jobs, and Scheduled Functions
Project: Automate daily report generation.
23. Day 23: Security Best Practices – Secrets, API Keys, RLS, and JWTs
Project: Secure a production application.
24. Day 24: Performance Optimization – Indexing, Query Optimization, and Connection Pooling
Project: Optimize a high-traffic application.
25. Day 25: Monitoring, Logging, and Debugging
Project: Diagnose and resolve production issues.
26. Day 26: Integrating Stripe, Webhooks, and Third-Party APIs
Project: Build a subscription billing system.
27. Day 27: Deploying Supabase Applications to Production
Project: Deploy a scalable SaaS platform.
28. Day 28: Multi-Tenant SaaS Architecture with Supabase
Project: Build a multi-organization project management platform.
29. Day 29: Scaling Supabase Applications – Replication, Caching, and High Availability
Project: Prepare an enterprise-ready backend.
30. Day 30: Final Capstone Project – Build a Production-Ready AI SaaS Platform with Authentication, PostgreSQL, RLS, Storage, Edge Functions, Real-Time Features, Vector Search, AI Agent Memory, Stripe Billing, Monitoring, and Cloud Deployment.
Grab the Supabase Mastery Ebook:
#Supabase# #PostgreSQL# #Backend# #Database# #AI# #ArtificialIntelligence# #GenerativeAI# #LLM# #RAG# #NextJS# #FullStack# #WebDevelopment# #Cloud# #SoftwareEngineering# #Programming#
Follow @e_opore to learn more.
Show more
AI coding agents might create an unexpected problem:
More code.
Less shared knowledge.
If an engineer solves a problem through an AI agent, a lot of the reasoning can stay inside a private human-agent loop.
No Stack Overflow answer.
No GitHub discussion.
No detailed issue thread.
Sometimes not even a meaningful commit message.
We may become dramatically better at producing software while becoming worse at producing the public knowledge that future engineers learn from.
That tradeoff deserves much more attention.
Show more
🚀 DeepSeek-V4-Flash Official API is now LIVE in public beta!
🔷 We’ve massively upgraded its Agent capabilities—benchmark scores are now far surpassing the V4-Pro-Preview. Check out the massive performance leap below! 👇
🔷 The official V4-Flash now natively supports the Responses API format and is fully adapted for Codex!
Check out the configuration details in our official API docs:
Show more
Kimi K3's KDA breaks conventional prefix caching: the recurrent state is cumulative, so per-block snapshots blow up memory. vLLM's fix — decoupling state-block size from prefix-match granularity — enables partial cache hits. Day-0 serving, ready before the Jul 27 weights. JP deep-dive 👇
Show more
今夜はマニアック回。昨日7/22、vLLMが「Kimi K3を公開初日から本番規模で配信するための裏側」を公開しました。
主題は、K3の新注意機構KDAが従来のprefix cachingの前提を壊す問題と、その解法。
↓ 推論基盤の人は必読です。
Show more
