AVAILABLE — IMMEDIATE JOINER · OPEN TO RELOCATE
Software Engineer | Java · Python · Full Stack & Generative AI | MCA 2026

Swapnil Panda
builds backend systems that ship.

Java (OOP) · Spring Boot · Microservices · Next.js · React.js · PostgreSQL · RAG & LLMs

MCA graduate specializing in Generative AI (SRM University, CGPA 9.20/10). Five months building production backend features at Plant Green Inertia, coupled with 300+ LeetCode DSA solutions and shipped full-stack systems spanning FinTech, clinical RAG healthcare, and AI task workflows. No toy projects — every system here is built, tested, and live.

View Projects & Live Demos Get In Touch Download Resume
📍 Chennai, Tamil Nadu, India +91 86581 19172 swapnilpanda26@gmail.com github.com/Swapnil26-art linkedin.com/in/swapnil-panda
9.20/10
CGPA · MCA GenAI (SRM)
300+
LeetCode DSA Solved
5 Mos
SDE Internship Experience
500+
Docs RAG Knowledge Base
OCI '25
Certified Data Science Pro
01 / About

Summary & Engineering Focus

MCA graduate specializing in Generative AI from SRM University (CGPA: 9.20/10, May 2026) with a strong academic foundation in Computer Science from Kalahandi University (B.Sc. Hons, CGPA: 7.04/10). Combines disciplined Object-Oriented software engineering in Java and Python with cutting-edge expertise in Generative AI, Retrieval-Augmented Generation (RAG), and modern full-stack application development.

Completed a 5-month Software Developer Internship at Plant Green Inertia Pvt. Ltd., Chennai, where I developed 8 RESTful Spring Boot/React.js API endpoints for internal reporting and engineered a Python/Flask microservice orchestrating external LLM APIs with custom prompt engineering to automate category summarization. Proven contributor across Agile sprint lifecycles with Git feature branching and thorough peer code reviews.

Demonstrates deep technical versatility through 300+ LeetCode Data Structures & Algorithms solutions and three production applications: a JWT-secured 3-tier Spring Boot backend (PayGenius), a real-time collaborative task platform powered by Next.js Server Actions and LLaMA 3.3-70B on Groq (TeamFlow), and an end-to-end clinical RAG chatbot indexed across 500+ medical sources with Pinecone and Azure OpenAI (Medical Chatbot). Immediate joiner and open to relocate for Software Engineer, Backend Developer, or Full Stack roles.

Backend & APIs

Java OOP, Spring Boot, Hibernate/JPA, REST API design, JWT security, Flask, and microservices architecture.

Generative AI & RAG

RAG pipelines, Pinecone vector indexing, LangChain, Hugging Face, Azure OpenAI, and Groq (LLaMA 3.3-70B).

Full Stack & DBs

Next.js Server Actions, React.js, PostgreSQL, MySQL, Prisma ORM, and responsive web systems.

Problem Solving & Cloud

300+ LeetCode DSA, Oracle Cloud (OCI 2025 Certified), Linux, Git, Postman, Docker, and CI/CD pipelines.

02 / Capabilities

Technical Skills

Languages

Java (OOP) Python JavaScript (ES6+) SQL HTML5 CSS3

Backend & APIs

Spring Boot Hibernate / JPA RESTful APIs JWT Auth Flask Microservices Maven

Frontend & Full Stack

Next.js (App Router) React.js Server Actions Responsive UI Tailwind CSS State Management

Databases & ORM

PostgreSQL MySQL Pinecone Vector DB Prisma ORM Spring Data JPA

Generative AI & LLMs

RAG Architecture LangChain Azure OpenAI Groq API (LLaMA 3.3-70B) Vector Search Prompt Engineering (Claude, ChatGPT, Copilot)

AI, ML & NLP

PyTorch TensorFlow scikit-learn Hugging Face NLP Deep Learning Fine-Tuning (Conceptual) Model Evaluation GridSearchCV

Cloud & DevOps

Microsoft Azure OCI (2025 Certified) Linux Git & GitHub Postman Docker CI/CD Pipelines

CS Core & Problem Solving

Data Structures & Algorithms 300+ LeetCode Solved OOP Design Principles Database Normalization Agile / Git Flow JUnit Redis
03 / Experience

Professional Experience

DEC 2025 — APR 2026 · 5 MONTHS
Software Developer Intern
Plant Green Inertia Pvt. Ltd. · Chennai, India
  • Situation/Task: The operations team was manually compiling weekly reports from scattered internal data, slowing turnaround. Action: Designed and shipped 8 RESTful API endpoints in Spring Boot and React.js for a new internal reporting module. Result: Report preparation moved from a manual process to a self-serve workflow for cross-functional teams.
  • Situation/Task: Report categorization still required manual review before distribution. Action: Built and deployed a standalone Python/Flask microservice that called external LLM APIs with targeted prompt engineering to auto-summarize 5 distinct report categories. Result: Cut manual review effort significantly while keeping summaries consistent across categories.
  • Situation/Task: The team needed stable, review-backed delivery across a 5-month Agile engagement. Action: Worked in Git feature branches with structured pull requests and peer code reviews, shipping on a regular sprint cadence. Result: Delivered regular releases over 5 months while keeping the production main branch regression-free.
  • Situation/Task: Reporting endpoints needed to fail safely under malformed or unexpected input. Action: Added backend validation and exception handling paired with matching React.js error states. Result: Secure, resilient payload handling across every endpoint with no unhandled failures in production.
JavaSpring BootPythonFlaskReact.jsREST APIsLLM IntegrationGitAgile
JUN 2025 — JUL 2025 · 2 MONTHS
Machine Learning Intern
Tamizhan Skills · Remote
  • Situation/Task: Needed a reliable way to compare model choices on structured data before committing to one. Action: Trained, evaluated, and benchmarked 5+ classifiers (SVM, Random Forest, Logistic Regression) on datasets of 5,000+ data points with cross-validation. Result: Delivered a clear, data-backed model comparison across classification and regression tasks.
  • Situation/Task: Baseline model accuracy was leaving performance on the table. Action: Ran systematic hyperparameter tuning with GridSearchCV and feature scaling over a 2-week Agile sprint, documenting the methodology throughout. Result: Improved model accuracy by 20% with a reproducible, documented process.
  • Situation/Task: Experiments needed to be consistent and repeatable across runs. Action: Built end-to-end data pipelines in Pandas and Scikit-learn covering missing-value imputation, feature scaling, and 80/20 train/test splits. Result: Consistent, reproducible benchmarking results across every experiment.
PythonScikit-learnPandasNumPyMachine LearningGridSearchCV
04 / Work

Featured Projects

Medical Chatbot
Live App Open Source
Aug 2025 — Nov 2025
AI Doctor Recommendation System — Retrieval-Augmented Generation (RAG) Healthcare Assistant
  • Situation/Task: General-purpose LLMs hallucinate on clinical queries without grounding in real medical literature. Action: Architected an end-to-end RAG pipeline over 500+ indexed medical documents in Pinecone. Result: Clinical queries return answers grounded in verified source text, eliminating ungrounded responses.
  • Situation/Task: Users needed symptom input to translate into an actual next step, not just an answer. Action: Unified Azure OpenAI for symptom extraction, Pinecone vector retrieval, and the Google Maps API for geolocation inside one Flask backend. Result: A single flow that turns symptoms into nearby specialist recommendations.
  • Situation/Task: Retrieval quality depends on how source text is chunked and embedded. Action: Used LangChain retrieval chains and Hugging Face embedding models with recursive character chunking. Result: Higher semantic precision and more relevant cosine-similarity matches at query time.
  • Situation/Task: Chained external API calls risk leaking secrets and stalling on slow services. Action: Isolated all API secrets in environment variables and orchestrated service calls asynchronously. Result: No timeout bottlenecks or integration failures across the live deployment.
Python Flask LangChain Azure OpenAI Pinecone Vector DB Hugging Face Google Maps API NLP JavaScript
Live Demo GitHub Repository
TeamFlow
Live App Open Source
Jan 2026 — Mar 2026
AI-Powered Collaborative Task Manager — Full-Stack Next.js with LLaMA 3.3-70B NLP
  • Situation/Task: Manually typing structured task fields slows teams down. Action: Built a full-stack Next.js task board that accepts conversational natural language, parsed and categorized by LLaMA 3.3-70B via the Groq API. Result: Tasks go from a plain sentence to a categorized board entry with no manual form-filling.
  • Situation/Task: The original LLM provider, Google Gemini, hit rate limits mid-project, threatening the delivery timeline. Action: Rapidly migrated the integration to Groq API (LLaMA 3.3-70B). Result: Eliminated dropped requests, achieved sub-second inference latency, and still shipped 100% of the planned scope on time.
  • Situation/Task: Task state needed to stay in sync across collaborators without friction. Action: Implemented Next.js Server Actions for live state sync and optimistic UI updates. Result: Tasks update instantly across concurrent sessions with zero manual page reloads.
  • Situation/Task: The board needed to stay fast and consistent as multiple users worked on it at once. Action: Designed normalized schemas in PostgreSQL via Prisma ORM with composite indexing on category and status filters. Result: Conflict-free performance verified across 10+ concurrent user sessions.
Next.js React.js Groq API (LLaMA 3.3-70B) PostgreSQL Prisma ORM Server Actions TypeScript Tailwind CSS
Live Demo GitHub Repository
PayGenius
3-Tier Backend Open Source
Dec 2025 — Feb 2026
FinTech Expense & Budget Tracker — Enterprise REST API Architecture
  • Situation/Task: A personal finance tracker needed a backend that could scale past a single-file prototype. Action: Architected a 3-tier backend in Java and Spring Boot with Controller, Service, and Repository layers across 4 domains: transactions, budgets, accounts, and reporting. Result: A clean, maintainable REST API structure ready to extend with new modules.
  • Situation/Task: Financial data demanded real access control, not just a login screen. Action: Enforced stateless authentication with Spring Security and JWT, BCrypt password hashing, and role-based endpoint guards. Result: Every endpoint is protected by verified, stateless auth with no plaintext credentials.
  • Situation/Task: Transaction, budget, and account data needed to stay consistent under validation rules. Action: Normalized schemas across 3 entity models in MySQL and PostgreSQL with tuned Hibernate/Spring Data JPA mappings and strict Jakarta validation. Result: Clean, validated data integrity enforced across 15+ REST endpoints.
  • Situation/Task: Manual testing wouldn't catch regressions as the API grew. Action: Built structured Postman test collections covering login flows, CRUD lifecycles, and edge-case validation failures. Result: Reliable, repeatable regression coverage throughout development.
Java (OOP) Spring Boot Spring Security JWT Hibernate Spring Data JPA MySQL PostgreSQL REST APIs Postman Maven
GitHub Repository
05 / Credentials

Education & Certifications

Master of Computer Applications (MCA) in Generative AI
SRM University, Chennai, Tamil Nadu
Jul 2024 — May 2026 · CGPA: 9.20 / 10

Specialized curriculum in Generative AI, Retrieval-Augmented Generation (RAG), Deep Learning, Large Language Models, Cloud Computing, and Advanced Data Structures & Algorithms.

Bachelor of Science (Honors) in Computer Science
Kalahandi University, Odisha
Aug 2021 — Apr 2024 · CGPA: 7.04 / 10

Core Computer Science focus: Object-Oriented Programming (Java/C++), Relational Database Management Systems (DBMS), Operating Systems, Computer Networks, and Software Engineering.

  • Oracle Cloud Infrastructure (OCI) 2025 Certified Data Science Professional
    Oracle · 2025 · Verified Credential
  • Career Essentials in Generative AI
    Microsoft & LinkedIn Learning · 2024
  • 300+ LeetCode DSA Problems Solved
    Demonstrated problem-solving proficiency across Arrays, Linked Lists, Trees, Graphs, Dynamic Programming, and Search/Sort in Java & Python
  • Cybersecurity CTF Hackathon
    Penetration testing, network vulnerability assessment & packet inspection using Nmap and Wireshark
06 / Contact

Let's build something.

Immediate joiner, open to relocate. Reach out for Software Engineer, Backend Developer, Full Stack, or AI Engineer opportunities.

swapnilpanda26@gmail.com +91 86581 19172 LinkedIn GitHub Resume