MinTax.
An AI document-classification workflow we co-own with a UK tax adviser: expense data in, HMRC-rule classification out, a PSA-ready analysis at the end.
Read the case study →Bravko is an automation engineering studio. We build AI workflows, browser automation, and data pipelines for e-commerce teams that are done with manual data work.
Fixed prices. Two-week pilots. You keep the code.
Every one of these is a pipeline we have built before.
Messy data in, clean structured output out. Document processing, catalog enrichment, classification and extraction pipelines built on LLMs, with validation and QA so the output is trustworthy enough to act on. Not chatbots. Working pipelines.
Price and competitor intelligence, market monitoring, product feeds, scheduled collection at scale. Built with Playwright and Python, engineered to survive site changes, and pointed only at data we can collect compliantly.
Automation is not a launch, it is an operation. We run what we build: uptime and drift monitoring, alerting, fixes and iteration on a monthly retainer, so it still works in month six.
An AI document-classification workflow we co-own with a UK tax adviser: expense data in, HMRC-rule classification out, a PSA-ready analysis at the end.
Read the case study →Everything above runs on the same stack we ship to clients.
Two weeks, fixed £1,900. We pick one painful process and automate it end to end. Example: daily price tracking on your top 100 products, report in your inbox every morning by 8am. 50% to start, 50% on delivery. You keep the system either way.
£4,000 to £12,000 depending on scope. The full pipeline: integrations with your stack, dashboards, alerting, documentation, handover.
£400 to £1,200 per month. We operate, monitor and improve what we built. Cancel monthly, no lock-in.
Most clients start with the pilot. Small enough to approve this week, real enough to prove the value.
No account managers, no handoffs, no juniors hiding behind a brand.
Error handling, retries, monitoring and alerts ship with every project, because automation that silently breaks is worse than none.
Everything lands in your repos and, where possible, your infrastructure. Leaving us is easy, which is why clients stay.
We are a remote studio on UK hours, and our cost base is not Shoreditch.
Overflow capacity on your paper. Senior Playwright, Python and LLM pipeline work, delivered white-label, day rate or fixed bid. If a build is stuck waiting on capacity, that is our favourite kind of email: info@bravko.com
Bravko is run by Mostafa Badr, a senior full-stack engineer who has spent 9 years building scraping systems, automation pipelines and web platforms for clients across Europe. He also runs his own production platform with thousands of daily users, built and operated end to end. He writes the code on your project.
Self-taught since his teens, shipping ever since.
We work with publicly available data, follow a compliance checklist per source, and design collection to be respectful: rate limits, no login walls, no personal data without a lawful basis. If a request crosses a line, we say no and propose an alternative.
Cairo, working UK hours. Contracts and invoicing run through Bravko LLC, payments by bank transfer or card. Time zone overlap with the UK is a full working day.
Python, Playwright, FastAPI, Postgres, and the LLM provider that fits the job. Deployed to your cloud or ours. Code delivered to your repositories with documentation.
The price was fixed, the scope was written down, and you keep whatever was built. No retainer obligation, no hard feelings. Our bet is that you will want step two.
Least-privilege access, NDAs as standard, read-only credentials wherever possible, and EU data stays where your compliance needs it.
A pilot usually starts within one week of the call.
Email us the most repetitive data job your team does. We will reply within one working day with how we would automate it, what it costs, and how long it takes. If it is not worth automating, we will tell you that too.
or write to info@bravko.com