Backend Engineer at Synexian (2025-10 – Present)
Sole backend engineer on two production applications using FastAPI, SQLAlchemy, Alembic, and Pydantic, owning end-to-end API architecture, service layer logic, schema modeling, database migrations, and deployment.
- Sole backend engineer on two production applications using FastAPI, SQLAlchemy, Alembic, and Pydantic, owning end-to-end API architecture, service layer logic, schema modeling, database migrations, and deployment.
- Architected a fine-grained User-Based Access Control (UBAC) system giving admins granular control over feature access, CRUD operation permissions, and data visibility per user.
- Produced a Markdown-compatible content management system supporting dynamic rendering and media handling.
- Assembled a bulk data ingestion pipeline processing 10,000+ records in under 60 seconds using Celery background workers, staging tables, row-level validation, partial failure handling (invalid rows rejected, valid rows committed), and duplicate control, spanning patients, treatments, invoices, dental services, and inventory.
- Formulated a modular backend architecture with clear separation of concerns to support scalability and maintainability.
- Enforced single-device login via a session control mechanism for improved account security.
- Developed backend support for analytics dashboards, including dynamic chart data and configurable visibility rules.
- Implemented a credit notes module from scratch integrated with the invoicing system to support refund workflows at a production level.
- Managed AWS infrastructure across two projects: Lightsail and EC2 (compute), RDS (database), S3 / AWS Object Storage (media and file storage), and Secrets Manager (secure configuration), with instances isolated from persistent storage.
- Containerized services using Docker and deployed a multi-service environment for shared testing and staging access.
- Personally migrated hosting from Render to AWS Lightsail, reconfiguring infrastructure, environment secrets, and deployment pipelines end-to-end.
- Wrote a shell script to fully automate deployment updates, requiring only GitHub credentials and AWS IAM details as input, eliminating manual release steps.
- Structured all backend APIs to support React frontend consumption, ensuring clean separation between frontend and backend layers.
Machine Learning Intern at Take It Smart (2023-10 – 2023-11)
Trained and deployed machine learning models using PyTorch and processed large datasets through ETL pipelines.
- Trained and deployed machine learning models using PyTorch, improving predictive accuracy on test datasets.
- Processed and validated large datasets through ETL pipelines, preparing structured inputs that improved model training efficiency.
