Building production-grade SaaS products, AI-powered systems, and automation workflows
I'm a full-stack developer and AI engineer based in π΅π° Pakistan, with 1.5+ years of experience shipping complete, production-grade SaaS applications from the ground up.
I specialize in integrating AI and automation into real products β not demos. My work spans the entire stack: frontend, backend, database design, AI pipelines, background jobs, and payment systems β always with clean architecture and maintainability in mind.
What I focus on:
- π€ AI-powered features β resume analysis, ticket routing, semantic search, chatbots
- β‘ Automation workflows with n8n, Inngest, and LangChain
- π Custom auth systems with token-based architecture (no third-party lock-in)
- π³ Credit-based payment systems with Stripe
- π Vector database integration (Qdrant) for semantic search and AI memory
- π¦ REST APIs with Swagger docs and production-level validation
Open to: Freelance Β· Contract Β· Full-time
A high-performance developer portfolio with Sanity CMS for dynamic blog content, achieving near-perfect Google Lighthouse scores across the board.
Highlights
- π 100% SEO Β· 100% Accessibility Β· 100% Best Practices Β· 98% Performance on mobile
- π― Perfect 100% across all metrics on desktop
- Sanity headless CMS for dynamic blog content
- Optimized for Core Web Vitals and page speed
Stack: Next.js Β· TypeScript Β· Sanity CMS Β· Tailwind CSS
A production-grade support platform that auto-analyzes incoming tickets, enriches them with AI metadata, and routes each one to the most relevant moderator using semantic search.
Highlights
- AI ticket enrichment + intelligent moderator routing via semantic search
- Qdrant vector DB for finding similar historical tickets instantly
- Role-based access control β admin, moderator, user
- Async processing via Inngest β background jobs never block the main request
- Admin panel with AI token usage and cost tracking, full activity logs
Stack: Next.js Β· Node.js Β· LangChain Β· Qdrant Β· PostgreSQL Β· Drizzle ORM Β· Inngest
An AI-powered platform where users upload their resume and receive a detailed, section-by-section breakdown with scores, identified weaknesses, and actionable improvement guidance.
Highlights
- Resume upload β AI-driven section-by-section analysis & overall scoring
- Credit-based billing via Stripe with full transaction history
- Custom auth system (no third-party providers) + full account management
- All analysis history persisted per user in MongoDB
Stack: Next.js Β· TypeScript Β· LangChain Β· OpenAI Β· MongoDB Β· Stripe
A modular, plug-and-play chatbot engine that handles all the complex backend aspects of a chatbot β drop it into any Node.js application in minutes.
Highlights
- Conversation memory + full session lifecycle management
- Tool calling support + real-time streaming responses
- Input validation with Zod, structured logging with Winston
- Fully modular β easy to extend and customize
Stack: Node.js Β· TypeScript Β· Express Β· LangChain Β· OpenAI
A YouTube-style backend platform built for scalable media workflows, multi-file uploads, and a fully documented REST API.
Highlights
- Multi-file upload for videos & thumbnails via Cloudinary
- Access + refresh token authentication
- Full Swagger API documentation
- MongoDB for flexible video, user, and metadata management
Stack: Node.js Β· TypeScript Β· Express Β· MongoDB Β· Cloudinary Β· Swagger Β· Zod
A production-grade LMS built for a real institute, fully digitalizing their academic operations from course management to student reporting.
Highlights
- Complete student, instructor, and admin workflows
- Course management, enrollment, assessments, and progress reporting
- Role-based access with secure data isolation per role
- Actively used by a local institute to manage academic operations
Stack: Laravel Β· MySQL Β· PHP Β· Blade
A complete, production-ready authentication foundation β no third-party providers, fully customizable, built to reuse across any SaaS project.
Highlights
- Full flows: sign-up, sign-in, email verification, password reset, forgot password
- Access + refresh token architecture with auto-regeneration on expiry
- Zero dependency on third-party auth providers
- Plug-and-play structure ready to drop into any Next.js SaaS
Stack: Next.js Β· TypeScript Β· JWT Β· Custom Token Auth
| π Portfolio | talhabilal.dev |
| πΌ LinkedIn | linkedin.com/in/talhabilal-dev |
| π¦ X (Twitter) | @talhabilaldev |
| π§ Email | contact@talhabilal.dev |
| π Location | Pakistan (PKT β UTC+5) |
"Build systems, not just features."









