Halo, Saya Deb 👋
Senior Fullstack & AI Engineer
Membangun sistem web interaktif performa tinggi dengan fondasi React.js, Vue.js, arsitektur database PostgreSQL, integrasi kecerdasan buatan (AI / LLM Pipelines), dan komputasi serverless ultra-cepat di Cloudflare Workers.
curl deb.sh
{ "frontend": ["React 19", "Vue 3", "Tailwind", "Next.js"], "database": ["PostgreSQL", "pgvector", "D1 SQL"], "ai_infra": ["Workers AI", "RAG Pipeline", "LLM Inference"], "security": ["Turnstile", "HSTS", "Strict CSP", "WAF"], "status": "Available for Projects" }
import { Pool } from 'pg'; import { embed } from 'ai'; // Cloudflare Hyperdrive + HNSW pgvector export async function semanticSearch(query: string, env: Env) { const pool = new Pool({ connectionString: env.HYPERDRIVE.connectionString }); const { embedding } = await embed({ model: embeddingModel, value: query }); return await pool.query( `SELECT id, title, 1 - (embedding <=> $1::vector) AS similarity ` + `FROM knowledge_base WHERE 1 - (embedding <=> $1::vector) > 0.85 ` + `ORDER BY embedding <=> $1::vector LIMIT 5`, [JSON.stringify(embedding)] ); }
Live Cloudflare Edge Diagnostics ONLINE
Telemetri real-time yang dihasilkan langsung dari Cloudflare Edge Worker terdekat.
Tentang Saya
Keahlian Rekayasa Modern & AI
Saya berdedikasi menciptakan aplikasi web modern dengan kombinasi UI reaktif (React.js & Vue.js), ketangguhan database relasional PostgreSQL, pipeline kecerdasan buatan cerdas, dan arsitektur serverless edge yang cepat dan scalable.
Reactive Frontend
Pengalaman pengguna responsif dengan React.js, Vue 3, Tailwind CSS, dan clean state management.
PostgreSQL & Vector Data
Desain skema terstruktur, indexing canggih, dan pgvector untuk pencarian semantik berkecepatan tinggi.
AI & LLM Workflows
Integrasi AI SDK, tool-calling otomatis, Retrieval-Augmented Generation (RAG), dan agentic workflow.
Edge Serverless Scaling
Cloudflare Workers untuk latensi ultra rendah, zero cold-start, dan proteksi keamanan di edge.
import { Pool } from 'pg';
import { embed } from 'ai';
// Accelerated Edge Inference with Hyperdrive & pgvector
export class EdgeAiService {
private readonly db: Pool;
constructor(env: Env) {
this.db = new Pool({ connectionString: env.HYPERDRIVE.connectionString });
}
async semanticSearch(query: string) {
const { embedding } = await embed({ model: embeddingModel, value: query });
return this.db.query(
`SELECT id, title, 1 - (embedding <=> $1::vector) AS similarity ` +
`FROM knowledge_base WHERE 1 - (embedding <=> $1::vector) > 0.85 ` +
`ORDER BY embedding <=> $1::vector LIMIT 5`,
[JSON.stringify(embedding)]
);
}
}
Tech Stack
Teknologi & Tooling Unggulan
Frontend Core
- React.js & Next.js SSR/Edge Runtime
- Vue.js 3 & Nuxt Composition & Nitro
- TypeScript Strict Type Generics
- Tailwind CSS Design Systems
AI & Machine Learning
- AI SDK & Tool-Calling Tool Calling & JSON
- pgvector & Embeddings HNSW Cosine Tuning
- RAG Pipelines Hybrid Search & Rerank
- Cloudflare Workers AI Serverless LLM
Database & Backend
- PostgreSQL (Postgres) HNSW pgvector & Tuning
- Redis (Cache/Queue) Distributed PubSub
- Node.js / Express / Hono Hono Edge Microservices
- Prisma / Drizzle ORM Type-safe Migrations
Cloud & DevOps
- Cloudflare Workers Hyperdrive & Caching
- Docker & Linux Multi-stage Builds
- GitHub CI/CD Actions Automated Edge CD
- REST & GraphQL APIs Contract OpenAPI
Showcase
Proyek & Karya Pilihan
Perjalanan Karier
Pengalaman Kerja & Kontribusi
Hubungi Saya
Mari Berkolaborasi
Punya ide proyek, kebutuhan sistem web modern dengan React/Vue, database PostgreSQL, integrasi model AI, atau konsultasi edge cloud? Kirimkan pesan Anda melalui formulir ini.