Software Engineer - Backend & RAG AI

Anjana Vollala

**************@*****.*** +1-682-***-**** TX, USA LinkedIn GitHub

Summary

Software Engineer with 3+ years of experience architecting scalable backend systems and production-grade AI solutions across regulated domains. Specialized in Python, FastAPI, React, GraphQL, and cloud-native platforms on AWS and Azure to deliver resilient, high-throughput services. Proven track record building fraud detection, NLP, and Retrieval-Augmented Generation applications that enhanced decision accuracy and operational efficiency. Adept at transforming complex data into deployable platforms, optimizing performance, and driving measurable outcomes in fast-paced engineering environments. Technical Skills

• Programming Languages: Python, JavaScript, TypeScript, SQL, Go

• Frontend Engineering: React, Angular, Next.js, Redux Toolkit, TanStack Query, HTML5, CSS3, Tailwind CSS, Progressive Web Apps, Server-Side Rendering, Web Accessibility (WCAG), Component-Driven Development

• Backend & API Development: FastAPI, Django, Flask, Node.js, RESTful APIs, GraphQL, gRPC, WebSockets, OAuth 2.0, JWT, Event- Driven Architecture, Async Processing, API Gateway, Serverless Architecture

• AI & Machine Learning: PyTorch, TensorFlow, Scikit-learn, Hugging Face Transformers, LangChain, LlamaIndex, Retrieval- Augmented Generation (RAG), Prompt Engineering, NLP, Recommendation Systems, MLflow, Kubeflow, AI Guardrails, Model Serving, Feature Engineering

• Cloud & DevOps: AWS (EKS, ECS, Lambda, SageMaker, S3), Microsoft Azure (AKS, Azure ML, Functions, Cosmos DB, Blob Storage), Docker, Kubernetes, Terraform, GitHub Actions, CI/CD Pipelines, Secure Coding, Scalability Practices

• Data & Platform Engineering: PostgreSQL, Redis, Apache Kafka, RabbitMQ, Elasticsearch, Pinecone, Weaviate, FAISS, Prometheus, Grafana, OpenTelemetry, PyTest, Jest, Performance Engineering, Distributed Caching Professional Experience

Software Engineer, PwC Jan 2025 – Present Remote, USA

Designed regulatory analytics platform on FastAPI and React with PostgreSQL, translating audit data requirements into scalable services that standardized compliance reporting workflows across multi-client financial engagements.

Implemented GraphQL services within Docker-based containerized architecture on AWS ECS, establishing data exchange patterns that strengthened integrations and shortened feature delivery cycles through GitHub Actions pipelines.

Developed fraud detection models using PyTorch on transaction streams, engineering behavioral features that elevated anomaly identification precision by 26% and enabled auditors to prioritize investigations with stronger risk signals.

Built Retrieval-Augmented Generation assistant using LlamaIndex and OpenAI embeddings to interpret policy documents, reducing research effort by 41% while improving interpretive consistency during time-sensitive audit reviews.

Operationalized recommendation models via Kubeflow pipelines, introducing automated retraining that increased cross-sell identification accuracy by 19% and supported advisory teams with opportunities across portfolios.

Established model observability with Evidently AI and Prometheus, surfacing drift insights that drove recalibration cycles and improved forecast stability by 17% across quarterly financial risk assessments. Software Engineer, TCS Aug 2021 – Jul 2023 Remote, India

Engineered claims processing platform using Python, Django, Angular, and MySQL, consolidating insurer workflows into unified services that reduced manual interventions and improved processing turnaround across distributed environments.

Architected event-driven microservices with Kafka and deployed containers on Kubernetes via Azure DevOps CI/CD pipelines, stabilizing release cycles and lowering production defects through automated testing and progressive rollout strategies.

Optimized relational queries and caching strategies within MySQL-backed APIs, accelerating response times for high-volume policy searches while supporting concurrent users without service degradation during regulatory reporting periods.

Built predictive claim severity model using Scikit-learn and feature-engineered historical adjudication datasets, enabling risk- based triaging that improved adjuster prioritization accuracy by 22% and shortened investigation queues.

Developed NLP pipeline leveraging Transformers to extract medical entities from unstructured documents, automating validation workflows and increasing straight-through claim eligibility detection by 31% across enterprise intake channels.

Implemented Retrieval-Augmented Generation framework with LangChain and vector indexing to assist agents with policy interpretation, decreasing knowledge lookup time by 37% and strengthening response consistency during audits.

Operationalized ML models through MLflow on Azure Kubernetes Service, introducing automated monitoring that detected drift early and sustained prediction reliability above 90% across quarterly model evaluations. Certifications

Machine Learning with Python: IBM Developers

Cloud Computing: IBM Developers

Backend Web Development with Node.js: UDEMY

Education

Master of Science in Computer Science, University of Texas at Arlington Aug 2023 – May 2025 Texas, USA Bachelor of Engineering in Computer Science and Engineering, MRIET Jul 2018 – Jul 2022 HYD, India

Contact this candidate