Skip to main content
Aarti Sharma

Aarti Sharma

Full-Stack Developer | Python | React | Node.js | AI/ML Integration

Talent
PuneMember since May 20, 2026
PythonTypeScriptSQLReact.jsNext.jsReduxFastAPIDjangoFlaskNode.jsExpress.jsRESTful APIsOAuth2RAG PipelinesVector DatabasesPineconeEmbeddingsPostgreSQLMongoDBGCPAWSDockerGitHub ActionsCI/CD PipelinesCloud RunArtifact RegistryMicroservicesWorkflow AutomationAsync ProcessingRBAC
·

Hire this person

Send a job offer directly to this candidate

PythonTypeScriptSQLReact.jsNext.js

About

Full Stack Developer with 2+ years of experience building scalable SaaS platforms, AI-powered applications, and cloud-native systems using Python, FastAPI, Django, Flask, React.js, Next.js, Node.js, and TypeScript. Experienced in backend engineering, REST API development, microservices architecture, OCR pipelines, Retrieval-Augmented Generation (RAG), vector databases, embeddings, and asynchronous processing systems. Hands-on expertise with GCP, Docker, CI/CD pipelines, GitHub Actions, Cloud Run, PostgreSQL, MongoDB, RabbitMQ, RBAC, OAuth2 authentication, passwordless authentication and workflow automation.

Strong experience working in Agile environments to deliver production-grade scalable applications.

Experience

Software Developer - La Net Team Software Solutions Pvt. Ltd. - Surat

(2024-07)

  • Built and shipped a production AI chatbot marketplace with React.js and Django, scaling to 500+ active users across 3 paid subscription tiers and driving recurring SaaS revenue.
  • Designed and deployed RAG pipelines using vector embeddings and semantic similarity search, reducing irrelevant AI responses by ~40% — improving user retention metrics across all subscription plans.
  • Integrated Pinecone vector database for embedding retrieval, cutting document search latency by 35% versus prior keyword-based lookup.
  • Implemented Stripe payment gateway for subscription billing, supporting 3 plan tiers and automated recurring payment workflows.
  • Built CI/CD pipeline using Docker, GitHub Actions, Artifact Registry, and Cloud Run — reducing average release time from 25 minutes to 8 minutes and cutting manual deployment steps by 60%.
  • Redesigned PostgreSQL schema for multi-tenant chatbot operations, improving query performance by 25% under peak concurrent load.
  • Architected a multi-tenant loan processing backend serving 7 financial clients, handling 1,000+ loan applications per month across isolated tenant environments.
  • Built Flask-based microservices for loan validation and approval workflows, reducing decision cycle time by 30% through automated rule evaluation.
  • Improved OCR-driven document ingestion pipelines that extracted structured financial data from uploaded files at 95%+ field-level accuracy, eliminating manual data entry for ~80% of submissions.
  • Introduced RabbitMQ-based async processing for high-volume submission queues, tripling system throughput and eliminating bottlenecks during peak windows.
  • Designed RBAC across 4 permission tiers (viewer, loan officer, manager, admin) for multi-tenant access control, ensuring data isolation and audit-trail compliance across all 7 clients.
  • Delivered a real-time async job-tracking dashboard, reducing unresolved stuck jobs by 70% and cutting support escalations from operations teams.
  • Built a React.js + TypeScript + Redux frontend for a startup/investor community platform, supporting complex portfolio management workflows for 200+ profiles.
  • Expanded RESTful APIs with Node.js and Express.js for investor portfolio and deal-flow modules, maintaining stable performance across 15+ concurrent third-party integrations.
  • Redesigned MongoDB schema and indexing strategy, cutting API response latency by 45% under concurrent load.
  • Led infrastructure migration from AWS to GCP, reducing monthly cloud costs by 20% while improving horizontal scalability for the platform.

Software Developer Intern - La Net Team Software Solutions Pvt. Ltd. - Surat

(2024-01 - 2024-06)

  • Developed 3 Django/FastAPI REST API modules now running in production, supporting core backend workflows for the live platform.
  • Built a spam email detection system using ML text classification (Naive Bayes, SVM), achieving 94% F1 accuracy on held-out test data.
  • Shipped a Chrome extension integrating real-time Gmail spam detection with Google OAuth authentication — deployed internally and published to the Chrome Web Store.
  • Trained and benchmarked multiple ML classifiers, selecting the highest-performing model based on precision, recall, and F1 score across cross-validation folds.

Education

Master of Computer Applications - Computer Applications - Veer Narmad South Gujarat University (VNSGU) (2022 - 2024)

Bachelor of Science - Computer Science - Veer Narmad South Gujarat University (VNSGU) (2019 - 2022)

Skills

Full-Stack DeveloperPythonReactNode.jsAI/ML Integration

Reviews

Related