What we build in Python
- APIs. REST APIs in FastAPI or Django REST Framework, with authentication, permissions, rate limits, tests and documentation.
- Django applications. Multi-tenant SaaS backends, internal tools and admin dashboards your team can actually use.
- Background work. Celery and Redis for scheduled jobs, retries and long-running tasks.
- Real-time and streaming. Kafka event streams into PostgreSQL and OpenSearch, pushed to dashboards as they happen.
- Data pipelines and scraping. Playwright collectors, cleaning, deduplication and delivery into your CRM or database.
- AI features. Assistants on your documents, extraction from PDFs and voice, with OpenAI, Claude or Gemini behind your own API.
Django or FastAPI?
| Django (with DRF) | FastAPI | |
|---|---|---|
| Best for | Full applications with users, admin, permissions and many data models | Fast, focused APIs and services, async workloads, AI and streaming endpoints |
| Comes with | ORM, migrations, admin, auth | Type-checked request and response models, automatic API docs, async by default |
| Typical use with us | SaaS backends, internal tools, platforms | Model-serving APIs, event processing, integrations |
Many systems use both: Django for the product and admin, FastAPI for the high-throughput or AI services next to it. We pick per project and explain why in the scope.
Built in production
- PulseOps: FastAPI and Kafka streaming into PostgreSQL and OpenSearch, with Valkey caching and Server-Sent Events, for a sub-second live view of every deploy and incident.
- Verita AI: a voice-first Django platform where consultation notes write themselves from the conversation.
- Search intelligence pipeline: 2,046 queries across 27 countries, parsed and batch-processed into one queryable dataset.
- Invoice workflow: OCR ingestion, proof of delivery, exports, reconciliation and approvals.
How we keep a codebase maintainable
- Typed code, tests on the parts that carry money or data, and CI on every change.
- Migrations, settings and secrets handled the same way in every environment, deployed with Docker on AWS or GCP.
- Documentation and a runbook, so your next engineer is productive on day one.
Hire a FastAPI or Django developer, without the hiring
If you were about to hire a Python developer for a backend, an API or an AI feature, the fractional retainer gives you a senior engineer for dedicated days each month instead: no recruiting, no notice period, pause or cancel anytime (see fractional AI engineer vs a full-time hire). You work directly with Saif, in your repositories and your cloud, and he can start within one to two weeks. For a defined piece of work, a fixed-scope build is usually cheaper than either.
Two ways to work with us
| Engagement | Price | Good for |
|---|---|---|
| Fixed-scope build | From $6,000 (data and automation) or $12,000 (product MVP) | A defined backend, API or pipeline |
| Fractional AI and Backend Engineer | Monthly retainer, pause or cancel anytime | Ongoing features, a codebase you want looked after, or senior help next to your team |
Frequently asked questions
Should I use Django or FastAPI?
Django when you are building a full application with users, admin and many data models; FastAPI for fast, focused APIs, async work and AI or streaming services. Many products use both side by side.
Can you take over an existing Python codebase?
Yes. We start with a short review of the code, tests, deployment and risks, then either fix what blocks you or take on ongoing work through the fractional retainer.
Can I hire a Python developer part-time?
Yes. The Fractional AI and Backend Engineer retainer gives you dedicated senior engineering days each month with direct access to Saif, and you can pause or cancel anytime.
Do you work in our repositories and cloud?
Yes, that is the default. Code lives in your repositories and runs in your cloud account from day one.
Which Python stack do you use?
Python, FastAPI, Django and DRF, Celery, PostgreSQL, Redis, Kafka and OpenSearch, Playwright for scraping, deployed with Docker on AWS or GCP.