Meshal Gohar Khan
HomeAboutProjectsContact
Let's talk
MGK
Meshal Gohar Khan

I design and ship AI-powered products, LLM apps, RAG pipelines, voice assistants, and production APIs backed by solid full-stack engineering in Node.js, Python, React, Vue, Laravel and modern web technologies.

Navigate
  • About
  • Projects
  • Contact
Connect
  • Email
  • GitHub
  • LinkedIn
© 2026 Meshal Gohar Khan.
Based in Malaysia·Open to collaborate when the fit is right

AI & Full-Stack Engineer

I design and ship AI-powered products, LLM apps, RAG pipelines, voice assistants, and production APIs backed by solid full-stack engineering in Node.js, Python, React, Vue, Laravel and modern web technologies.

Approach, Discuss and Go Ahead

Start a conversationView work
Stack I use in production: OpenAI | RAG | Python | FastAPI | ChromaDB | Node.js | TypeScript | React | Vue | Next.js | Turborepo | PostgreSQL | MongoDB | Docker | Laravel
Meshal Gohar Khan. Profile photo
Code with me

About

AI systems, backed by full-stack craft.

I'm an AI & Full-Stack Engineer with 5+ years of experience, currently based in Malaysia.

My background spans backend architecture, API design, and frontend systems across logistics, oil & gas, e-commerce, and enterprise software. I work primarily with Node.js, Python, Laravel, React, Vue, and PostgreSQL/MongoDB/MySQL and I've been shipping production AI features with OpenAI, Whisper, ElevenLabs, and retrieval-augmented systems.

My current focus is AI engineering: LLM-powered products, RAG / document intelligence, conversational and voice interfaces, and agentic workflows that stay grounded, evaluable, and production-ready.

If a collaboration or role is a strong fit, I'm open to remote, hybrid, or onsite arrangements, including relocation and visa sponsorship where needed.

I work asynchronously well, communicate clearly in English, and have experience across distributed cross-timezone teams.

Tech

Stack

AI / GenAI
OpenAI APIsLLM ApplicationsPrompt EngineeringAI IntegrationsConversational AIWhisper (STT)ElevenLabs (TTS)AI PrototypingRAG / Document IntelligenceSemantic EmbeddingsAgentic Workflow Exploration
Backend Engineering
API DesignNode.jsNestJSExpressStripeLaravelPHPPythonFastAPIGraphQLWebSocketsZodAxiosPusherMongoosePinoJWT
Frontend Engineering
TypeScriptJavaScriptReact.jsNext.jsVue.js (Vue 2/3)ViteTailwind CSSBootstrapPiniaChart.jsPlotlyFramer MotionVuex
Databases
MySQLMongoDBPostgreSQLChromaDBFirebase Firestore
Infrastructure
TurborepoDockerCI/CDAWS S3FirebaseGitPostmanLinuxStructured LoggingSystem Design Basics

Focus

AI Systems Focus

  • RAG & document intelligence
  • LLM-powered applications
  • Grounded retrieval & source attribution
  • Conversational & voice AI
  • Prompt orchestration & evaluation
  • Agentic systems exploration
  • API-integrated AI products
  • Dockerized AI services

Research

AI Safety

Beyond engineering: I'm deeply interested in the reliability, alignment, and evaluation challenges that emerge when LLMs are deployed in real-world production environments. My production experience with LLM failure modes, prompt brittleness, hallucination, output evaluation gaps, has shaped a serious interest in AI safety research.

Career

Professional Experience

Shipping production software with ownership across AI features, backend, frontend, delivery, and mentoring.

2021

Present

Mar 2021 — Present

Senior Software Engineer

Synesis Labs (formerly AlphaSquared) · Full Time · Remote

At Synesis Labs I grew from Full-Stack Developer to Senior Software Engineer by consistently delivering production-ready features across logistics, analytics, and AI-assisted platforms.

On the backend I designed and built secure RESTful APIs using Node.js, NestJS, Laravel and Python with a strong focus on performance, access control, and maintainability. One of my key wins was improving API response times by up to 40% through targeted query optimization and database indexing across live production systems.

I built and integrated AI-powered features using OpenAI APIs, conversational workflows, prompt orchestration, voice transcription via Whisper, and synthesized audio responses via ElevenLabs, shipped to real users in customer-facing applications.

On the frontend I built scalable, reusable component systems in React and Vue, integrated with high performance backend APIs and real-time event workflows using WebSockets and Pusher.

I also owned CI/CD pipelines using GitHub Actions and Docker, conducted regular code reviews, mentored junior engineers, and contributed to architecture decisions across the team.

2019

2021

Nov 2019 — Feb 2021

Laravel Developer

NetRoots Technologies LLC · Full Time · Onsite

At Netroots Technologies I worked as a Laravel Developer building and maintaining enterprise business automation platforms, including HRM, ERP, and CRM systems used across HR, finance, procurement, and inventory operations.

I built backend modules from scratch and enhanced existing ones to improve operational efficiency, and integrated third-party services including payment gateways, cloud APIs, and notification systems.

I was involved end-to-end, from development and testing through to deployment and collaborated on technical documentation that supported ongoing implementation and QA processes. This role gave me a strong foundation in PHP, Laravel architecture, production debugging, and building systems that real businesses depend on daily.

Case studies

Production & client work

Flagship delivery plus supporting systems, each linking to live surfaces where available.

  • Case studyFull-Stack2023 — present

    FluidsData

    A characterization data solution for the oil and gas industry, offering end-to-end services from data digitization and ingestion to AI-ML model development. The platform streamlines data standardization, enhances visualization, democratizes data access, and enables accurate fluids property predictions, significantly reducing sampling and analysis costs. With unmatched expertise in both fluid characterization and data science, FluidsData empowers clients to maximize the value of their proprietary data assets while retaining full ownership.

    • Vue.js
    • Python
    • Plotly
    • Data Visualization
    Live product →
  • Case studyPlatform Engineering2021 - present

    Who I Follow

    This application integrates e-commerce with social networking for sports enthusiasts. Users can create accounts, make friends, share posts, and view profiles. It features individual and team chats, event and game creation. Invitations are sent via sms and email. Shop includes browsing, cart management, and Stripe payments, all managed through an Admin Panel overseeing ads and analytics. The app supports game scoring, tournament management, and role-based administration within teams.

    • Laravel
    • Vue.js
    • Stripe
    • MySQL
    Live product →
  • Case studyFull-Stack2025 - 2026

    Craftsman Web

    Craftsman Web is a comprehensive field service management platform designed to streamline the operations of Craftsman Furniture, a leading provider of furniture solutions. The platform provides a range of features to manage bookings, reports, and support for clients, as well as jobs, technicians, clients, invoicing, and route planning for admins. The platform is backed by an Express API with authentication, job/report data, and optional Google Places for reviews.

    • React.js
    • Node.js
    • MongoDB
    Live product →
All projects

Engineering

Selected Engineering Projects

AI-first personal builds, RAG, voice, chat, and more, with source on GitHub and live demos where available.

  • AI Document IntelligenceIn progress2026/mesh-rag

    Mesh RAG

    A full-stack Retrieval-Augmented Generation platform for intelligent document question answering. Next.js drives PDF upload and conversational querying, while a FastAPI service owns the retrieval pipeline: text extraction, chunking, semantic embeddings, ChromaDB vector search, prompt construction, and grounded LLM responses with source attribution.

    • Next.js
    • TypeScript
    • FastAPI
    • Python
    • ChromaDB
    • OpenAI
    • RAG
    Request details ->GitHub
  • AI + Full-StackLive demo2026/meshai-support

    MeshAI Support

    MeshAI Support is a full-stack chat workspace backed by MongoDB and OpenAI. Operators or end users work in a responsive React shell; the Express API isolates routing, controllers, and services, exposes a small REST surface, and is packaged for traditional Node hosting or a hosted HTTP worker using the same codebase.

    • React
    • Vite
    • Tailwind CSS
    • Node.js
    • Express
    • MongoDB
    • OpenAI API
    Live demo →GitHub
  • AI + VoiceLive demo2026/meshvoice-ai

    MeshVoice AI

    MeshVoice AI is a voice-driven assistant: users record in the browser, the backend transcribes audio, generates a short spoken-style reply with an LLM, and returns synthesized audio for playback. The stack is React and Vite on the client, Express on the server, with OpenAI and ElevenLabs as external providers.

    • React
    • Vite
    • Node.js
    • Express
    • OpenAI
    • ElevenLabs
    Live demo →GitHub
  • Turborepo MonorepoLive demo2026/mesh-portfolio

    Mesh Portfolio

    This portfolio is a Turborepo monorepo: Next.js apps, shared UI/config packages, Mongo-backed visitor analytics, and admin tooling in a dark-first marketing site.

    • Turborepo
    • Next.js
    • TypeScript
    • Tailwind CSS
    • MongoDB
    • Framer Motion
    Live demo →
  • Data VisualizationLive demo2026/meshanalytics-dashboard

    MeshAnalytics Dashboard

    A production-grade SaaS analytics dashboard that visualizes revenue, pipeline, channel performance, and account-level insights for go-to-market teams. Built as a clean monorepo with a Vue 3 client and an Express API designed for cloud deployment.

    • Vue 3
    • Pinia
    • Chart.js
    • Vite
    • Node.js
    • Express
    • Zod
    • Axios
    Live demo →GitHub
  • Multi-tenant SaaSLive demo2026/meshcore

    MeshCore

    Multi-tenant control-plane style API with an admin SPA. The repo is a small monorepo - Express backend, Vite + React frontend, and static hosting that routes /api to the same Express app. MeshCore demonstrates a production-shaped split between HTTP routing, controllers, domain services, and infrastructure (PostgreSQL, structured logging, centralized errors), with a signed-in workspace for tenant-scoped lists and read-only analytics cards.

    • TypeScript
    • React
    • Vite
    • Express
    • PostgreSQL
    • Zod
    • Axios
    • Pino
    Live demo →GitHub
  • Realtime SystemsLive demo2026/meshchat-realtime

    MeshChat Realtime

    Production-oriented group chat with JWT auth, MongoDB persistence, and Pusher Channels for messaging, typing, and presence. MeshChat is a full-stack reference: a Vite React client and an Express API for traditional or cloud hosting; realtime goes through Pusher while the API owns authorization, validation, and database writes.

    • React
    • Vite
    • Node.js
    • Express
    • MongoDB
    • Mongoose
    • Pusher
    • JWT
    Live demo →GitHub
All projects

Now

Currently building

Hands-on AI systems I own end-to-end, from retrieval and model orchestration to the product surface.

  • Building

    RAG & document intelligence

    End-to-end retrieval pipelines: PDF ingestion, chunking, embeddings, ChromaDB vector search, and grounded LLM answers with source attribution.

    →In motion
  • Shipping

    LLM apps & voice AI

    Conversational support tools, Whisper transcription, ElevenLabs speech, and prompt orchestration shipped as real product surfaces.

    →In motion
  • Iterating

    AI workflow tooling

    Backend services and developer-facing systems that wrap LLMs behind structured APIs, evaluation hooks, and production-ready architecture.

    →In motion

Contact

Let's build something intelligent.

Tell me about the AI surface, constraints, and what “done” looks like, I'll reply with honest feasibility and next steps.

Prefer async? Drop a message, same inbox as serious threads.

Email directlyOpen full contact