I am a Software Engineer and M.S. Software Engineering student at Arizona State University building applied AI tools, backend/cloud systems, and Android product workflows.

I like building software that is reliable, explainable, and useful in real-world workflows. My projects connect backend services, cloud infrastructure, mobile systems, and AI-assisted automation.

I build AI-assisted tools that combine retrieval, model reasoning, validation, and structured outputs. I am interested in RAG pipelines, verifier workflows, local evaluation, model explainability, and systems where AI is used as one controlled part of a larger software workflow.

I build APIs, services, and data workflows with a focus on clean architecture, reliability, and performance. I am interested in service design, authentication, event-driven systems, API gateways, async processing, and backend workflows that are easy to test and maintain.

I work with cloud-native architectures using AWS services such as Lambda, API Gateway, DynamoDB, S3, and EC2. I am interested in serverless systems, deployment automation, infrastructure documentation, CI/CD pipelines, cost-aware design, and production-style cloud workflows.

I have experience building Android and mobile-integrated systems involving Kotlin, Jetpack Compose, Room, foreground services, background tasks, notification listeners, device-context signals, Firebase, REST backend integration, and permission-aware user flows.

Some of my current work focuses on explainability, privacy, user control, and safe automation. I am interested in systems that make intelligent decisions while avoiding unnecessary exposure of private data.

Agentic AI contract analysis system for risk detection, evidence scoring, verifier review, and clause rewrite generation.

• Built a multi-stage legal document analysis pipeline with preprocessing, local RAG, compliance checking, verifier review, weighted scoring, clause rewriting, and Markdown/JSON report generation
• Added deterministic mock-model execution and benchmark workflows for reproducible demos
• Included tests, validation scripts, and clear limitations for responsible AI use

Serverless user management and authentication system using Spring Boot and AWS.

• Built RESTful user-management APIs with Spring Boot, AWS Lambda, API Gateway, DynamoDB, and CloudFormation
• Added infrastructure-as-code templates, API documentation, deployment guide, and testing structure
• Focused on cloud-native architecture, scalability, and serverless deployment patterns

Cloud-native microservices backend for e-commerce workflows.

• Built modular services for authentication, users, products, carts, orders, orchestration, discovery, and API gateway routing
• Used Spring Boot, Kafka, Eureka, JWT, MySQL/JPA, and distributed architecture patterns
• Designed for service separation, event-driven communication, async workflows, and scalable backend design

Research engineering contribution to an Android and Flask-based platform for privacy-aware mobile context modeling, backend ML lifecycle management, and server-side LLM-assisted user-facing outputs.

• Built Android/Kotlin workflows using Jetpack Compose, Room, DataStore and shared preferences, coroutines, foreground/background services, notification-listener integration, permission handling, and device-context readers
• Integrated a Flask backend with participant-scoped data storage, context ingestion, feedback-label ingestion, model lifecycle checks, health/config endpoints, and diagnostic responses
• Supported backend ML pipelines with pandas, scikit-learn, joblib, ensemble models, confidence outputs, influential-feature reporting, SHAP-style explanations, and retraining hooks
• Added server-side LLM guardrails including provider/mode validation, mock fallback, prompt/output validation, retry logic, API key isolation, OpenRouter budget checks, and safe fallback behavior
• Wrote tests for server flow, participant isolation, model readiness, LLM fallback, budget handling, Android policy gates, and system-event eligibility

Custom multilingual conversational LLM training pipeline.

• Built a data pipeline for WhatsApp export parsing, cleaning, tokenization, training, evaluation, and packaging
• Implemented a custom tokenizer and lightweight transformer workflow for Hindi-English code-mixed conversational data
• Focused on privacy-filtered data handling, reproducibility, and small-model limitations

• Building applied AI systems with retrieval, verifier workflows, explainability, and structured outputs
• Building backend and cloud-native systems with stronger deployment, reliability, and observability practices
• Developing privacy-aware Android and backend ML systems with mobile context modeling, server-side LLM components, diagnostics, and user-control safeguards
• Actively interviewing for software engineering internship and new-grad opportunities across applied AI, backend systems, cloud engineering, full-stack engineering, and Android/mobile development

• Portfolio: arpit-jaiswal.vercel.app
• GitHub: arpitJ-dev
• LinkedIn: linkedin.com/in/arpitj16
• Email: arpitjaiswal.dev@gmail.com