Skip to content
Yogith Ramanan

Open to internships & full-time opportunities

Yogith Ramanan

Computer Science Graduate Student · Software Engineer · Cloud · Backend · AI/ML · DevOps

Cloud-native systems, production-grade APIs, ML pipelines, and serverless architectures — engineered for reliability, scale, and real-world impact.

I'm an M.S. Computer Science candidate at DePaul University, specializing in software engineering, cloud engineering, backend development, full-stack development, machine learning, and DevOps. I've shipped production features, resolved critical incidents, and built AWS-native pipelines, REST APIs, and CI/CD workflows through internships and academic projects. My current focus is research on carbon-aware scheduling for multi-region AI inference. I'm driven by scalable architecture, observability, automated testing, reproducible workflows, and clear cross-functional communication.

View projectsDownload ResumeContact me
Yogith Ramanan, professional portrait

Profile

About me

Education, technical focus, and the engineering mindset behind every project.

I'm pursuing an M.S. in Computer Science at DePaul University, building on my B.E. in Computer Science from Sri Krishna College of Engineering and Technology. My academic path reflects a strong foundation in software engineering, now extended with graduate-level work in cloud systems, backend architecture, applied machine learning, and DevOps pipelines.

I'm motivated by roles where reliability, scalability, and clarity matter — designing and operating backend services and APIs, shipping cloud-native and serverless systems, and applying machine learning where it delivers measurable outcomes. I'm comfortable working across the full stack when needed, with particular interest in integration, observability, and systems that teams can test, monitor, and improve with confidence.

My internship experience in product development support and full-stack delivery taught me how production issues surface for real users and how disciplined debugging, clear documentation, and proactive communication accelerate fixes. I bring a consistent bias toward automation, testing, and reproducible workflows — habits I've sharpened through coursework, hands-on projects, and research.

Campus photo

  • Graduate study

    M.S. Computer Science

    DePaul University · Chicago, IL · 2025–present

  • Undergraduate

    B.E. Computer Science

    Sri Krishna College of Engineering and Technology · 2020–2024

  • Professional focus

    Software · Cloud · Backend · AI/ML

    Full-stack delivery, cloud-native systems, applied ML & research

Work

Experience

Hands-on experience delivering production-grade software, supporting live systems, and building end-to-end features under real-world constraints.

  1. Product Development and Support Engineer Intern

    January 2024 – June 2024

    GURU Information Technology Services

    Chennai, Tamil Nadu, India

    • HTML / CSS / JavaScript
    • Log analysis
    • MTTR
    • Cross-functional collaboration
    • Supported product stability for customer-facing applications by resolving UI defects and production incidents across HTML, CSS, and JavaScript — reducing user-reported issues and improving end-user experience for enterprise clients.
    • Accelerated incident resolution through log analysis, targeted patches, and documented remediation — minimizing repeat occurrences and reducing mean time to recovery (MTTR) across production systems.
    • Authored detailed defect reports and validation summaries using structured debugging workflows; collaborated cross-functionally with engineering teams to enforce quality gates and improve deployment readiness end-to-end.

  2. Full Stack Intern

    July 2022 – September 2022

    Gateway Software Solution

    Coimbatore, Tamil Nadu, India

    • Full-stack
    • Dashboards
    • GitHub workflows
    • Client delivery
    • Contributed to full-stack product delivery for a software consultancy, building interactive dashboards and data-driven interfaces that translated backend data into actionable business insights for end users.
    • Maintained front-end, back-end, and database layers in parallel — reducing regressions across releases and ensuring consistent feature delivery against evolving client requirements.
    • Applied GitHub-based version control workflows for branching, code review, and team collaboration; enforced deployment readiness standards that improved review efficiency and reduced integration conflicts.

Research

Featured research

Research in cloud-scale AI inference, carbon-aware scheduling, and simulation-driven evaluation for sustainable infrastructure.

Research highlight

Carbon-Aware Scheduling for Multi-Region AI Inference

January 2026 – March 2026

  • Python
  • Cloud Systems
  • Simulation

Research on routing AI inference workloads across global cloud regions when both carbon intensity and latency constraints must be optimized simultaneously. Built a discrete-event simulator on real-world traces, evaluated hybrid routing policies against static baselines, and produced an enterprise decision framework for sustainable multi-region deployment.

View paper (PDF)Repository

Key outcomes

  • Discrete-event simulation evaluating 8 routing policies across 5 AWS regions and 3 AI workloads (BERT-base, BERT-large, ResNet-50) — 33,600 requests over 7 days with real carbon intensity and latency data.
  • Constrained Hybrid policy: 54.8% per-request carbon cost reduction with only 24.8 ms average latency overhead and 0% SLO violations — deployable in cloud load balancers.
  • Adaptive Hybrid controller using closed-loop P95 latency feedback — 29.7% carbon reduction at 4.53% SLO violation rate without manual parameter tuning.
  • Enterprise decision framework mapping 5 deployment scenarios (strict SLO through batch async) with Pareto trade-off analysis across all 8 policy configurations.

Builds

Projects

Selected projects demonstrating cloud engineering, applied machine learning, full-stack development, and production-ready system design.

AgentMind — Autonomous Multi-Agent AutoML System preview

Screenshot placeholder. Replace the file at /images/projects/agentmind.jpg or point imageSrc to your asset in src/data/portfolio.ts.

Featured project

AgentMind — Autonomous Multi-Agent AutoML System

April 2026 – June 2026

Free open-source AutoML platform targeting the $1.1B+ market as an alternative to DataRobot ($2,500/month) — enabling business analysts and enterprises to obtain a trained, optimized ML model in under 60 seconds with no data science expertise required.

  • LangGraph
  • FastAPI
  • React
  • Groq
  • ChromaDB
  • Vite
  • scikit-learn
  • XGBoost
  • +6 more

Highlights

  • 88.5% accuracy on heart disease prediction and 85.3% on a 145,460-row weather dataset — compressing days of model selection into under 60 seconds with full SHAP explainability.
  • Parallel ML pipeline training 5 models simultaneously (Random Forest, XGBoost, LightGBM, CatBoost, Logistic Regression) via ThreadPoolExecutor with Optuna Bayesian tuning.
  • Full-stack React + Vite frontend with FastAPI backend — deployed live on Vercel and Render with CI/CD auto-deploy on every GitHub push.
Live demoRepository
  • Multimodal AI Platform for Drug Candidate Prioritization preview

    Screenshot placeholder. Replace the file at /images/projects/drug-prioritization.jpg or point imageSrc to your asset in src/data/portfolio.ts.

    Multimodal AI Platform for Drug Candidate Prioritization

    January 2026 – March 2026

    ML-driven drug candidate screening platform addressing pharmaceutical R&D bottlenecks where discovery costs exceed $2.5B per approved compound — offering reproducible, cost-effective prioritization as an alternative to manual assay workflows.

    • PyTorch
    • FastAPI
    • Docker
    • MLflow
    • RDKit
    • scikit-learn
    • +3 more

    Highlights

    • Graph-tabular fusion model outperformed logistic regression baseline and Random Forest fusion across ROC-AUC, PR-AUC, F1, and rank correlation — demonstrating that molecular graph structure improves prioritization quality.
    • Deterministic candidate ranking with rule-based descriptor penalties, confidence scoring, and reason codes for auditable decision support.

    +1 more highlight

    DemoRepository
  • Serverless Document Processing System preview

    Screenshot placeholder. Replace the file at /images/projects/serverless-docs.jpg or point imageSrc to your asset in src/data/portfolio.ts.

    Serverless Document Processing System

    September 2025 – October 2025

    Fully serverless, event-driven document digitization pipeline on AWS — replacing manual OCR workflows with zero-ops processing that scales from 1 to millions of files without dedicated infrastructure overhead.

    • AWS Lambda
    • S3
    • Textract
    • DynamoDB
    • API Gateway
    • CloudWatch
    • +2 more

    Highlights

    • End-to-end processing from S3 upload to structured text extraction and queryable API — automatic scaling with pay-per-use Lambda pricing and near-zero operational cost.
    • RESTful API Gateway endpoints with API-key authentication, least-privilege IAM controls, and structured 4xx/5xx lifecycle handling.

    +1 more highlight

    DemoRepository
  • MealSnap AI — Image-Based Food Calorie & Nutrition Tracker preview

    Screenshot placeholder. Replace the file at /images/projects/food-calorie.jpg or point imageSrc to your asset in src/data/portfolio.ts.

    MealSnap AI — Image-Based Food Calorie & Nutrition Tracker

    June 2025 – August 2025

    Production ML app targeting the $220B+ digital health and nutrition market — deployed as a live public web app with real-time food recognition, calorie estimation, and nutritional tracking for consumer wellness use cases.

    • TensorFlow
    • MobileNetV2
    • Streamlit
    • Python
    • Food-101
    • USDA FoodData Central

    Highlights

    • 85.3% top-1 and 96.2% top-3 accuracy on 101 food classes at under 250 ms CPU inference latency.
    • Multi-signal OOD detection system (confidence threshold, entropy, top-3 spread) achieving 95%+ non-food rejection for reliable daily nutrition decisions.

    +1 more highlight

    Live demoRepository
  • A2A Java SDK — Agent-to-Agent Communication Platform preview

    Screenshot placeholder. Replace the file at /images/projects/a2a-sdk.jpg or point imageSrc to your asset in src/data/portfolio.ts.

    A2A Java SDK — Agent-to-Agent Communication Platform

    January 2025 – March 2025

    Strongly typed Java SDK for enterprise agentic AI architectures — reducing integration time for multi-agent systems by providing reusable client abstractions that replace ad-hoc HTTP wiring with standardized, observable inter-agent communication.

    • Java
    • Spring Boot
    • REST APIs
    • Maven
    • JUnit

    Highlights

    • Strongly typed, reusable SDK replacing ad-hoc HTTP wiring — cutting boilerplate, enforcing consistent error contracts, and enabling traceable inter-agent calls with zero additional instrumentation.
    • Rigorous payload validation, edge-case handling, and standardized error handling following SOLID principles.

    +1 more highlight

    DemoRepository

Academics

Education

Formal training that underpins software, systems, cloud, AI/ML, and DevOps work.

  • Master of Science in Computer Science

    DePaul University

    Chicago, IL, USA

    January 2025 – Present

  • Bachelor of Engineering in Computer Science

    Sri Krishna College of Engineering and Technology

    Coimbatore, Tamil Nadu, India

    July 2020 – May 2024

Stack

Technical skills

Organized by category to match my resume - everything listed below appears there; nothing is trimmed for effect.

  • Languages

    • Java
    • Python
    • SQL
    • JavaScript
    • Bash

  • Frontend / UI

    • React
    • Vite
    • HTML5
    • CSS3
    • Streamlit

  • Backend / APIs

    • FastAPI
    • Flask
    • Django
    • Spring Boot
    • Node.js
    • REST APIs
    • OpenAPI
    • GraphQL

  • Cloud (AWS)

    • EC2
    • S3
    • Lambda
    • RDS
    • API Gateway
    • CloudWatch
    • SageMaker
    • EKS
    • IAM
    • VPC
    • Textract
    • EventBridge
    • AWS Batch

  • DevOps / MLOps

    • Docker
    • Docker Compose
    • Kubernetes
    • Terraform
    • GitHub Actions
    • CI/CD
    • Jenkins
    • MLflow
    • Linux

  • AI / ML

    • PyTorch
    • TensorFlow
    • scikit-learn
    • XGBoost
    • LightGBM
    • CatBoost
    • SHAP
    • Optuna
    • Bayesian Tuning
    • Feature Engineering
    • Transfer Learning
    • GCN
    • Multimodal AI
    • Parallel Model Training

  • Agentic AI

    • LangGraph
    • LangChain
    • Multi-Agent Orchestration
    • RAG
    • ChromaDB
    • Groq API
    • OpenAI API
    • Prompt Engineering
    • Chain-of-Thought Reasoning

  • Data / Databases

    • PostgreSQL
    • MySQL
    • MongoDB
    • DynamoDB
    • ChromaDB
    • Pandas
    • NumPy
    • Spark

  • Cheminformatics

    • RDKit
    • SMILES
    • Morgan Fingerprints
    • Molecular Descriptors
    • Drug Candidate Prioritization

  • Observability / Testing

    • Prometheus
    • CloudWatch
    • pytest
    • JUnit
    • Structured Logging
    • Integration Testing

  • Principles

    • OOP
    • SOLID
    • Design Patterns
    • Agile/Scrum
    • System Design
    • Microservices
    • Distributed Systems

Resume

Quick highlights & download

At-a-glance signals for recruiters; explore projects and experience above for depth.

Graduate study

M.S. in Computer Science

DePaul University · 2025–present · Chicago, IL

Internships

Two software engineering internships

Product development & support (GURU IT) · full-stack delivery (Gateway Software)

Build & ship

Cloud, backend, and applied ML projects

AWS serverless pipelines, APIs, SDK work, and multimodal ML platforms

Research

Simulation-driven, sustainability-aware systems

Carbon-aware multi-region inference routing — evaluation & decision frameworks

Full resume

Download a single PDF with consolidated experience, education, and skills—easy to share with recruiters and hiring managers.

Download Resume
GitHubLinkedIn

Reach out

Contact

I’m actively seeking internships and full-time opportunities in Software Engineering, SDE, Cloud/DevOps, Backend Development, and AI/ML roles. I’m especially interested in building scalable systems, cloud-native platforms, reliable backend services, automation pipelines, and applied machine learning solutions that create real-world impact.

If you’re hiring or want to discuss a project, reach me by email or the form—LinkedIn works too.

Send a message

Submitting opens your email app with a pre-filled message—no server storage. You can also email me directly at yogithramana@gmail.com.