[01 / PROFILE_SUMMARY]
I am JA Shuvro (legally MD. Jonaed Ali Shuvro), an Applied AI Engineer & Full-Stack Developer with 3.5+ years of active system development. My core capability lies in architecting production-grade Applied AI systems—multi-modal LLM pipelines (OpenAI Whisper, GPT-4o, TTS), vector search (RAG & Assistants API), and asynchronous evaluation queues—integrated with robust enterprise web and mobile runtimes (Next.js, NestJS, WordPress, Flutter, PostgreSQL).
[03 / DEPLOYED_SYSTEMS]
PROJECT 01: HR Interview System — AI-Powered Video & Voice Mock Interview Platform
Role: Applied AI Engineer & Full-Stack Architect. Orchestrated multi-modal AI pipelines utilizing OpenAI Whisper for speech-to-text audio transcription, OpenAI TTS for voice question synthesis, and GPT-4o for rubric scoring. Cut initial screening time by 75% with a 98.5% transcription accuracy.
Technologies: WordPress Plugin, React 18, OpenAI Whisper, GPT-4o, OpenAI TTS, WebRTC, PHP 8, REST API, OneCloud Credits.
Links:[Case Study](GitHub: Private Repository)
PROJECT 02: Medical Interview Bot — Clinical AI Mock Interviewer & Diagnostics Viva Simulator
Role: Applied AI Engineer & Clinical System Integrator. Developed a multi-instance WordPress CPT architecture allowing medical institutions to deploy specialty viva examiners. Ingests medical CVs and case documents (PDF, DOCX, Excel) to generate differential diagnosis questions via GPT-5-Nano / GPT-4o-mini in under 2.4s.
Technologies: WordPress Plugin, OpenAI GPT-5-Nano, GPT-4o-mini, Whisper STT, OpenAI TTS, smalot/pdf-parser, PhpWord, PhpSpreadsheet, WebRTC.
Links:[Case Study](GitHub: Private Repository)
PROJECT 03: Mentoro (Study Mentor) — Decoupled AI Study Mentor & Learning Pack Engine
Role: Applied AI Engineer & Distributed Systems Architect. Built a decoupled AI educational system combining a NestJS microservice backend with a WordPress React SPA Question Player. Ingests lecture PDFs, Word documents, and YouTube lecture transcripts to generate flashcards, quizzes, and automated rubric grading for student answers in under 8s.
Technologies: NestJS, TypeScript, WordPress Plugin, React SPA, OpenAI GPT-4o, mammoth, officeparser, pdf-parse, youtube-transcript, Tailwind CSS.
Links:[Case Study](GitHub: Private Repository)
PROJECT 04: WP AI Tools Suite — Multi-Model AI Productivity Suite & Vector Search RAG
Role: Applied AI Engineer & Multi-Model Integrator. Consolidated 6 distinct AI capabilities into a WordPress React 18 SPA: OpenAI Assistants API vector store search (RAG), Chat Completions, DALL-E 3 image generation, Whisper STT, TTS, and RunwayML video generation, backed by a three-tier token credit management engine saving 40% in API costs.
Technologies: WordPress Plugin, React 18, Mantine UI, OpenAI Assistants API (Vector Stores), GPT-4o, DALL-E 3, Whisper, OpenAI TTS, RunwayML.
Links:[Case Study](GitHub: Private Repository)
PROJECT 05: Flirtmetrics — Real-Time Chat & Matchmaking Mobile App
Role: Mobile Specialist & Flutter Lead. Optimized UI card swiping to achieve a stable 60 FPS by isolating Riverpod state rebuilds. Resolved 8-second chat delivery latency issues by transitioning the application to event-driven WebSockets and local SQLite cache.
Technologies: Flutter, Riverpod, WebSockets, SQLite, Firebase, REST.
Links:[Case Study][Play Store][App Store](GitHub: Private Repository)
PROJECT 06: Enterprise ERP System — Supply Chain Ledger Platform
Role: Database Architect & Full-Stack Developer. Transformed manual operations (Excel ledgers, paper signatures) into a real-time ledger synchronization pipeline. Reduced inventory discrepancies to <1% and cut approval delays from 4 days to under 30 minutes.
Technologies: Next.js, React, Node.js, PostgreSQL, Tailwind CSS.
Links:[Case Study][Live Demo](GitHub: Private Repository)
PROJECT 07: National Environmental GIS Monitoring Platform (NEGMP)
Role: System Architect & Proposal Author. Designed a conceptual GIS monitoring platform proposal for a national 250M tree plantation program. Prepared a universal GeoEntity spatial layout, an offline-friendly Flutter design for low-connectivity pilot zones, and a conceptual Google Earth Engine satellite verification workflow.
Technologies: NestJS, Next.js, Flutter, PostgreSQL + PostGIS, Google Earth Engine API, Keycloak, Leaflet.js / MapLibre GL, Redis, Prisma.
Links:[Case Study](GitHub: Under Review - MoEFCC)