About
We build data pipelines.
Creative Codes is a software studio. We scrape data, process it with ML, and automate what happens next.
What we do
I've spent years building each of these individually: scrapers for e-commerce and intelligence work, ML models for classification and extraction, automation pipelines connecting CRMs to data feeds. At some point the pattern was obvious. They're not three services. They're three stages of the same pipeline. Building them together from the start is the only way to avoid the integration tax that kills every data project.
Clients are product teams, ops leads, and research groups who need data moving reliably from the web into their systems. They don't want to build the infrastructure themselves, or maintain it when it breaks.
What we build with
Backend
Data
Infrastructure
Founder
Muhammad Hassan
Founder & Lead Engineer
8 years building backends. The last 3 spent going deep on AI agents, RAG pipelines, and production scraping systems. The stack I work in daily: Python, LangChain, LangGraph, Playwright, n8n, FastAPI, Qdrant, and whatever the problem actually requires.
The work spans a few recurring categories: RAG pipelines for companies that have thousands of internal documents and need accurate answers from them, scraping systems that collect structured data at scale from targets that actively resist it, voice AI platforms for telecom and real estate, and automation pipelines that eliminate ops work across CRMs, support queues, and data workflows. Clients are across the US, UK, EU, Middle East, and Australia — mostly B2B SaaS, e-commerce, finance, and real estate.
Most of the early projects came through Upwork, where I still take on work. Clients kept coming back with bigger problems, which is how Creative Codes became a studio instead of a freelance profile.
I write the code, architect the systems, and get on the calls. No account managers, no project coordinators. You talk to the person building your system. That means slower sales but faster delivery — there's no translation layer between what you need and what gets built.
Every project is fixed-price. That removes the incentive to over-engineer or drag out timelines. Code goes into your GitHub from day one — no lock-in, no black box, no dependency on us to make changes after handoff.
How we work
We measure what we ship, we deploy to production (not just staging), and we own the full system from scraper to output. If a proof-of-concept doesn't lead to a production deployment, we consider it a failure.
Location
Based in Faisalabad, Pakistan. Our working hours overlap with US East and UK business hours. Most communication happens async through Loom and Slack.
Available for calls between 6 PM - 2 AM PKT (9 AM - 5 PM EST)
Want to work together?