The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
I build backend systems that stay reliable as products grow.
Over the past few years, I’ve worked on SaaS platforms, payment systems, healthcare products, marketplaces, logistics applications, AI-powered tools, and cloud infrastructure.
My work has included:
• Scaling APIs handling millions of requests per day
• Building payment and subscription workflows
• Designing PostgreSQL and MySQL database architectures
• Developing AI applications, RAG systems, and document-processing pipelines
• Building event-driven systems with queues and background workers
• Deploying applications using AWS, Cloudflare, and Docker
• Improving API performance, reliability, monitoring, and security
My main stack includes Node.js, TypeScript, Python, PostgreSQL, Redis, RabbitMQ, Cloudflare Workers, AWS, Docker, Stripe, and OpenAI.
I’m especially interested in working with startups that need help building a new product, improving an existing backend, integrating payments or AI, or preparing their platform for growth.
Open to backend development, AI engineering, architecture consulting, and long-term product collaborations.
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Related posts
A duplicate webhook should not change inventory twice.
Commerce Sync is my multi-store Shopify integration example. Product changes move between storefronts, and inventory updates make their way back to the parent store.
The part I care about is what happens when something fails. PostgreSQL catches duplicate deliveries, pg-boss keeps jobs durable, and separate workers handle throttling and retries. Partial syncs have a rollback path.
I put the architecture and implementation decisions in the case study below. Client details are removed, and the render is a concept illustration.
I am taking on Shopify app integrations and backend maintenance, remotely from India, up to 20 hours a week.
A great breakdown of the modern AI ecosystem! 🚀 Building robust, scalable AI applications requires bringing together the right tools at every layer, from vector databases for context to evaluation frameworks for reliability.
A great breakdown of the modern AI ecosystem! 🚀 Building robust, scalable AI applications requires bringing together the right tools at every layer, from vector databases for context to evaluation frameworks for reliability.
I've been heavily focused on working across this...
My Role & Contributions
I served as the Lead Product Manager for TalentScreen Pro, an AI-powered recruitment and candidate-assessment platform.
I was responsible for taking the product from concept to development and commercialisation.
My key contributions included:
Leading product planning, strategy and development.
Defining product requirements and managing the product roadmap.
Coordinating the engineering team and sprint delivery.
Developing AI-powered interview generation and video-screening workflows.
Building candidate assessment processes to support recruitment.
Overseeing product delivery and commercialisation.
Project Outcome:
TalentScreen Pro was successfully developed and commercialised, generating reported sales of US$10,000.
The project demonstrated my ability to turn an idea into a functional AI-powered SaaS product.
Getting an AI screening tool to real revenue is the hard part. How did you handle recruiter pushback on bias and transparency in the video screening flow? That is usually where these products stall.
