# Al Amin Ahamed > Senior Software Engineer — LLM Integration & Agentic Systems — Building production LLM integration pipelines, RAG architectures, and agentic systems — shipped inside WordPress plugins, Python services, and Claude Code tooling. Anthropic Certified. WordPress Polyglots Translation Editor. Based in Dhaka. - GitHub: https://github.com/mralaminahamed - LinkedIn: https://linkedin.com/in/mralaminahamed - Contact: me@alaminahamed.com ## Pages - [Home](https://alaminahamed.com/): Introduction, featured work, recent writing, and current status. - [About](https://alaminahamed.com/about): Background, skills, experience, and approach to software. - [Work](https://alaminahamed.com/work): Open-source projects and commercial work (45 projects). - [Writing](https://alaminahamed.com/writing): Technical articles (43 posts). - [Stack](https://alaminahamed.com/stack): Tools and technologies I use daily. - [Now](https://alaminahamed.com/now): What I'm working on and reading at the moment. - [Hire](https://alaminahamed.com/hire): Availability and engagement options. - [AI Chat](https://alaminahamed.com/ai): Ask questions about my work, projects, and writing. - [MCP Server](https://alaminahamed.com/mcp/portfolio): Remote MCP endpoint for AI assistants (OAuth 2.1 required). ## Writing 43 published articles. Topics: Laravel, PHP, AI/LLMs, WordPress, DevOps, React, TypeScript, performance, and security. Tags: AI / LLMs, Anthropic, Architecture, Caddy, Career, Cloudflare, Deployment, DevOps, Docker, Dokan, Gutenberg, Laravel, MySQL, Open Source, PHP, Performance, Pest, PostgreSQL, RAG, React, Redis, SQLite, Security, Tailwind CSS, Testing, TypeScript, Vite, WooCommerce, WordPress, pgvector ### Recent Articles - [Anthropic Academy — Six Certifications Completed](https://alaminahamed.com/writing/anthropic-academy-certifications) — I completed six Anthropic Academy courses covering the Claude API, Model Context Protocol, subagents, agentic workflows, Claude Code, and AI fluency — each with a verified certificate from Anthropic Education. - [Prompt Versioning in Production: Managing LLM Prompts Like Code](https://alaminahamed.com/writing/prompt-versioning-production-llm) — How EasyCommerce manages LLM prompts across Claude, OpenAI, and DeepSeek with a versioned registry, structured evals in CI, and a rollback path that does not require a deployment. - [RAG Without a Framework: A Minimal pgvector Pipeline with Claude](https://alaminahamed.com/writing/rag-without-framework-pgvector-claude) — How to build a complete RAG pipeline in Python without LangChain or LlamaIndex — chunking, embedding with text-embedding-3-large, pgvector retrieval, and grounded generation with Claude — with the tradeoffs that matter at production scale. - [Building a Claude Code MCP Server in Python: Lessons from codebase-research-agent](https://alaminahamed.com/writing/python-mcp-server-claude-code) — How codebase-research-agent exposes a tool-using AI agent as a Claude Code MCP server — JSON-RPC over Streamable HTTP, tool registration, streaming SSE, and the four things that broke in production before it worked reliably. - [How I Cut Edit Distance from 168 to 43 in a Legal Document RAG Pipeline](https://alaminahamed.com/writing/legal-rag-edit-distance) — Three architectural changes that brought average edit distance from 168 to 43 in a legal document RAG pipeline — section-scoped chunking, section-scoped exemplar prompting, and switching from a general embedding model to a legal-domain one — with a CI eval suite that made the improvement trajectory visible and reproducible. - [Designing a ReAct-Style Codebase Research Agent with Tool Use](https://alaminahamed.com/writing/codebase-research-agent-react) — How codebase-research-agent uses a ReAct loop to orchestrate semantic search, AST navigation, symbol lookup, grep, and git blame as callable tools — the hybrid retrieval substrate it operates on, and why framing RAG as a tool rather than a pipeline changes the architecture fundamentally. - [Building a 12-Source Job Discovery Pipeline with RAG and pgvector](https://alaminahamed.com/writing/jobpulse-rag-pipeline) — How JobPulse aggregates job listings across 12 sources in three tiers, embeds your resume into pgvector, and generates grounded cover letters via Claude — including the async adapter protocol, composite scoring model, and the three retrieval failure modes that shaped the final architecture. - [Scaling a Laravel MCP Tool Registry to 46 Tools](https://alaminahamed.com/writing/laravel-mcp-tool-registry-at-scale) — The May 12 post covered the MCP server architecture. This one goes inside the tool registry: class-per-tool structure, Sanctum ability scoping, JSON Schema validation before dispatch, tag-based auto-discovery, and the three things that broke when the registry hit 46 tools. - [Stop Using Fixed-Size Chunks for Technical Documentation](https://alaminahamed.com/writing/heading-aware-chunking-rag) — Why naive fixed-size chunking breaks on code-heavy documentation, and the heading-aware approach with paragraph fallback and overlap that replaced it in this portfolio's RAG pipeline. - [Voyage AI vs OpenAI Embeddings for Technical RAG in PHP](https://alaminahamed.com/writing/voyage-vs-openai-embeddings-rag-php) — Why I switched from OpenAI text-embedding-3-small to Voyage AI voyage-code-3 for this portfolio's RAG layer — model comparison, the input_type asymmetry Voyage requires, measured retrieval improvement, and the migration path that keeps search live throughout. ## Projects - [Codebase Research Agent — ReAct AI Agent (Claude Code MCP)](https://alaminahamed.com/work/codebase-research-agent) — A ReAct tool-using AI agent over hybrid code retrieval (pgvector + BM25 + AST), exposed as a Claude Code MCP server. - [JobPulse RAG — AI Job Discovery & Cover Letter Platform](https://alaminahamed.com/work/jobpulse-rag) — A 12-source job discovery platform with RAG-grounded, personalized cover letter generation — built on FastAPI, pgvector, and a multi-provider LLM layer. - [StoreSheet – WooCommerce Google Sheets Sync](https://alaminahamed.com/work/storesheet) — Two-way sync between WooCommerce and Google Sheets. Products, orders, coupons, customers, and variations — live in your spreadsheet. - [Medicare — Hospital Management System](https://alaminahamed.com/work/laravel-medicare-app) — Full-stack hospital management system — clinical workflows, IPD/OPD, pharmacy, lab, billing, patient portal, and Stripe memberships. Built with Laravel 13, React 19, and Inertia v3. - [Fashion Factory — Multi-tenant SaaS](https://alaminahamed.com/work/laravel-fashion-factory-saas) — Laravel SaaS platform for fashion factories and retail shops — multi-tenancy, production orders, inventory management, POS, and integrated accounting. - [PMS Aggregator — Property Management Dashboard](https://alaminahamed.com/work/laravel-pms-aggregator-app) — Laravel platform that aggregates property listings, bookings, and reporting from multiple PMS providers into a single unified dashboard. - [The WP Squad — Company Website](https://alaminahamed.com/work/the-wp-squad-website) — Marketing and product website for The WP Squad — a WordPress plugin company. Built with Laravel 11, Tailwind CSS v4, Alpine.js, and React. - [TubeGrab — TypeScript Desktop Downloader](https://alaminahamed.com/work/tubegrab-ts) — Cross-platform Electron desktop app for downloading YouTube videos and playlists — format selection, quality picker, progress queue, and yt-dlp backend. ## Experience - Senior Software Engineer at Codexpert, Inc (2026–present) - Team Lead of Product at Codexpert, Inc (2025–present) - Software Engineer at Codexpert, Inc (2025–2025) - Software Engineer at weDevs (2024–2025) - Software Engineer at Riseup Labs (2023–2024) - Software Engineer at EchoaSoft Ltd. (2023–2023) - Software QA Engineer at EchoaSoft Ltd. (2021–2023) ## Stack - **Framework:** FastAPI - **Languages:** PHP, TypeScript, Go, Python, Rust - **Frameworks:** Laravel, WordPress, React, Inertia.js, Livewire, Alpine.js, Tailwind CSS - **Databases:** PostgreSQL, MySQL / MariaDB, Redis, SQLite, Meilisearch - **AI / ML:** Voyage AI, LangChain / LlamaIndex, pgvector, Ollama, OpenAI API - **DevOps:** Docker, GitHub Actions, Caddy, Hetzner Cloud, Supervisor, Linux / Ubuntu - **Tools:** PhpStorm, WP-CLI, Pest, PHPStan, Laravel Pint, Vite, Playwright, TablePlus, Figma