build: passing page loads −35% deploy setup −60% ml accuracy 89% scheduler satisfaction 92% bugs −25% build: passing page loads −35% deploy setup −60% ml accuracy 89% scheduler satisfaction 92% bugs −25%

software engineer · bengaluru, in

Zainab Bi

I build full-stack apps
and applied ML models that works in production.

CS graduate · ex-intern at HostUp Cloud Technologies · Java, Spring Boot, React, Next.js, Node.js, Docker, cloud-native and a habit of measuring everything I ship.

I'm a Computer Science graduate (CGPA 8.5/10) from Atria Institute of Technology, with hands-on experience shipping cloud-native web applications during my internship at HostUp Cloud Technologies. I'm equally comfortable configuring VPCs and containers as I am training a classification model — my projects span full-stack platforms, DevOps tooling, and applied machine learning, usually all three at once.

Feb 2026 – May 2026
Screenshot of internship certificate

Software Engineer Intern

HostUp Cloud Technologies Pvt. Ltd. · Bengaluru

  • Developed cloud-native Next.js applications covering authentication, billing, and ticketing, cutting page load times by 35%.
  • Containerized deployments with Docker, reducing deployment bugs by 25%.
  • Configured cloud infrastructure — VPCs, subnets, security groups — cutting environment setup time by 60%.
  • Worked within DevOps practices that reduced post-deployment issues by 30%.
Screenshot of the glycemic control classifier output

Glycemic Control ML Classifier

89% accuracy

A Random Forest pipeline that classifies diabetes risk conditions, with feature selection tuned so predictions return in under 100ms — fast enough to sit behind a live product, not just a notebook.

PythonNumPyScikit-learnPandas
Screenshot of modern blog

Modern Blog Platform

~150–300ms response

A full-stack blog engine with Next.js and MongoDB (via Prisma), covering CRUD operations at 90–95% functional accuracy, built for fast expansion as features get added.

Next.jsNode.jsTypeScriptPrismaBun

Genetic Algorithm Class Scheduler

92% satisfaction

An AI-powered university timetable generator combining genetic algorithms with OR-Tools, serving optimized schedules in ~200–500ms per request with real-time updates and JWT-secured access.

FastAPIMongoDBReact/TSOR-ToolsGoogle Gemini
Languages
JavaJavaScriptTypeScriptPythonHTML5CSS3
Frameworks
React.jsNext.jsNode.jsExpress.jsTailwind CSSBootstrapHibernate
Data
MongoDBMySQLNoSQLPrisma ORM
Cloud & DevOps
DockerKubernetesAWSMicrosoft AzureApache KafkaApache SparkGitCI/CDLinux
Authentication
REST APIFast APIJWTOAuth
Fundamentals
DSAOOPSOperating SystemsMultithreading
Focus
AI/MLCyber SecurityData AnalyticsCloud ComputingComputer Networking

Certificates

Screenshot of cyber security certificate
Cyber Security — Deloitte
Screenshot of data analytics certificate
Data Analytics — Deloitte
Screenshot of java certificate
DSA with Java — Apna College
Screenshot of java certificate
Web Development — Apna College
Screenshot of python certificate
Python Foundation — Infosys

Achievements

Screenshot of publication
Co-authored Scopus-indexed blockchain voting research, presented at ICETSE 2025.
Screenshot of acceptance
Published AI research on diabetes classification in Gradiva Review Journal.

Let's build something that ships.