Connect the systems your business already runs and automate the work in between. WhatsApp Business API, Stripe, Twilio, QuickBooks, Airtable and Google Workspace.
The unglamorous half of software: two systems that both hold the truth about a customer, and disagree. Getting that right is mostly care - idempotency, retries, a mapping written down, and an alert when a run fails rather than a dashboard that has silently been stale for a fortnight.
I run these in my own product, which is a useful discipline. When a webhook double-fires at 2am it is my problem, not a client’s.
Authentication, the API calls, and the mapping between two systems that disagree about what a customer is.
Idempotency, so a duplicate delivery does not charge twice or create the record twice.
Retries with backoff, a dead-letter path, and an alert. A silent integration is a broken one you have not noticed yet.
Backfill as well as ongoing, because the first run is never the only run.
Most integration pain is not the API. It is that one system's order is another's invoice plus shipment, and nobody wrote down the mapping.
Then a small real run, watched, before anything is switched on fully.
Logs, a status you can look at, and an alert when a run fails.
WhatsApp Business API in production in my own product, plus Stripe, Twilio, QuickBooks, Airtable, Fireflies, Outlook and Google Workspace - and a Django application integrated with several third-party APIs at 5.0 stars on Upwork.
Sometimes, and I will say so. Zapier or Make is occasionally the right answer and cheaper than anything I would build. Once the logic turns conditional, or the volume gets real, hand-built wins.
Yes. I have built a QuickBooks-to-BigQuery sync in production, so I know where its API is awkward.
Retrieval-augmented AI that answers from your documents and cites the page it used, or says it does not know. Built and run in production, not a demo.
Move your data somewhere it can be asked questions, and build the dashboards on top. BigQuery, dbt, Airflow, Power BI and Looker Studio, shipped for clients since 2019.
Inherit, stabilise and keep running software someone else built. Bug fixes, upgrades, cloud costs, and an honest assessment of what you actually have.
The work behind these is public: the AI product, its live demo, the cited answer pages it produces, and the research paper measuring it.
Send a short brief — what you need built, roughly when, and the budget you have in mind. You'll get a reply from Muhammad Kashif Irshad himself, in writing, within one working day, saying plainly whether this is a good fit.
