Ritika Banerjee

Backend Engineer

Hyderabad, Telangana 1+ yrs exp 83 · Excellent

About

Engineer building production backend systems (FastAPI, PostgreSQL, async pipelines, real-time communication) for AI-powered applications, with end-to-end ownership from data ingestion to deployed inference. Combines backend architecture with applied computer vision/ML, shipping systems that are both scalable and model-aware.

Skills & Expertise (38)

FastAPI Advanced

8.0/10

Years Exp

REST APIs Intermediate

7.5/10

Years Exp

Postgresql Intermediate

7.5/10

Years Exp

Python Intermediate

7.5/10

Years Exp

SQL Intermediate

7.0/10

Years Exp

WebSockets Intermediate

7.0/10

Years Exp

Async Processing Intermediate

7.0/10

Years Exp

OpenCV Intermediate

7.0/10

Years Exp

Pandas Intermediate

7.0/10

Years Exp

NumPy Intermediate

7.0/10

Years Exp

Git Intermediate

7.0/10

Years Exp

Docker Intermediate

7.0/10

Years Exp

Plotly Intermediate

6.5/10

Years Exp

Matplotlib Intermediate

6.5/10

Years Exp

PGvector Intermediate

6.5/10

Years Exp

YOLOv8 Intermediate

6.5/10

Years Exp

TensorFlow Intermediate

6.0/10

Years Exp

Google Cloud Platform Intermediate

6.0/10

Years Exp

Pinecone PyTorch Keras RoboFlow Image Processing n8n CrewAI LangGraph LangChain ChromaDB CNNs scikit-learn Minio LINUX Azure Pipelines Object Detection Segmentation NLP Hugging Face Transfer Learning

Work Experience

Backend Engineer – Machine Learning & Computer Vision Systems

Meridian Data Labs

Apr 2026 - Present

Architected an end-to-end backend system integrating FastAPI, PostgreSQL/pgvector, and MinIO for automated image ingestion, storage, and visual retrieval from Raspberry Pi-based capture devices. Designed an asynchronous processing pipeline using TaskIQ to offload compute-intensive workloads (image processing, embedding generation) from the API request path, improving API responsiveness under load. Integrated SAM for ROI segmentation, pgvector for similarity search, and TripoSR for 3D model generation, connecting model outputs directly into the retrieval backend. Own the end-to-end backend for an industrial inspection platform, designing and deploying production services using FastAPI, Docker, and PostgreSQL. Built a WebSocket-based communication layer between two hardware devices to send and receive real-time camera on/off signals, enabling synchronized state tracking across distributed inspection hardware. Optimize object detection pipelines for manufacturing environments using advanced image preprocessing and post-processing techniques, improving model robustness and production reliability.

Machine Learning & Computer Vision Intern

Meridian Data Labs

Aug 2025 - Mar 2026

Architected and deployed the initial inspection pipeline backend using FastAPI and Docker, enabling containerized model inference, data processing, and API integration. Achieved 92% detection accuracy with a 15% improvement in minority-defect recall by optimizing the object detection pipeline using OpenCV, CLAHE, Gabor filtering, and adaptive masking.

AI Engineer (Freelancer)

Avlys AI

Jul 2025 - Oct 2025

Engineered a modular FastAPI backend with multi-platform publishing APIs, reducing manual campaign deployment by 70% while maintaining API response times below 500 ms. Built an end-to-end data pipeline aggregating business-relevant content from social platforms, news, and web sources, improving AI-generated content relevance by 30%+.

AI/ML Intern

Neurapses Technologies

Apr 2025 - Aug 2025

Improved industrial defect detection accuracy by 80% through advanced preprocessing and augmentation on 3,000+ raw jute images. Reduced inference time by 35% using a hybrid YOLOv8 segmentation and Swin Transformer classification pipeline optimized with ONNX.