Unwritten helps businesses scale by providing complete, integrated growth solutions without the need to manage multiple agencies. We grew from consulting roots after seeing how overpriced, non-transparent agency models limited client success. Today, Unwritten is a network of 20+ specialists in design, technology, marketing, and AI, delivering end-to-end solutions that bring marketing, sales, tech, and design together into a unified growth engine.
We're looking for a Python developer with 3+ years of experience to join us full-time, in a hybrid setup (3 days WFO initially - can be reduced depending on performance).
What You'll Work On
- Data pipelines (Python/FastAPI): Build and maintain services for PubMed article ingestion, categorization, translation, author-data migration, and URL enrichment — several of which run as independent Dockerized microservices.
- Search infrastructure: Design and optimize indexing pipelines and queries against OpenSearch, keeping article and video content searchable and performant at scale.
- AI-assisted content enhancement: Work with AWS Bedrock and Google Gemini (via LangChain) to enrich data, generate metadata, and power classification/relevance features (e.g., related-articles-by-topic).
- Data layer: Work with MySQL (via SQLAlchemy/aiomysql) for article, author, and citation data, and integrate with WordPress-based content APIs.
- Testing & CI: Write unit/integration tests (pytest, GitLab CI) and contribute to improving test coverage across pipeline services.
- Cloud & infra: Work with AWS (S3, Bedrock, IAM), Docker/Docker Compose for local dev and service packaging, and observability via OpenTelemetry.
Tech Stack You'll Use
- Languages: Python 3+
- Frameworks: FastAPI
- Data & Search: MySQL, OpenSearch, SQLAlchemy
- AI/ML: AWS Bedrock, Google Gemini, LangChain, scikit-learn
- Infra: Docker, Docker Compose, AWS (S3, IAM), GitLab CI, OpenTelemetry
- Tooling: Poetry, pytest
What We're Looking For
- 3+ years of backend development experience, with solid Python skills (bonus: FastAPI, async/await patterns with asyncio/aiohttp).
- Experience with relational databases (MySQL/PostgreSQL) and search engines (OpenSearch/Elasticsearch preferred).
- Familiarity with containerised development (Docker, Docker Compose) and CI/CD pipelines.
- Exposure to AWS services (S3, IAM, Bedrock or similar LLM/AI services) is a strong plus.
- Comfort working across a polyglot codebase with multiple independently deployable services.
- Strong testing habits and ability to write maintainable, well-tested code.
- Nice to Have
- Experience with LLM-based content pipelines (LangChain, prompt engineering, structured extraction).
- Familiarity with PubMed/NCBI E-utilities or scientific publishing metadata.
- Experience with data migration/ETL pipelines at scale.