Srivatsa Koustub

About

Results-driven Backend & Machine Learning Engineer with 1.10+ years of hands-on experience in Python, Flask, Django, and REST API development. Proven expertise in building production-grade backend systems, implementing DevOps practices across full development cycles, and building and integrating AI/ML capabilities into production backend systems. Exceptional written and verbal communication skills with confident presentation abilities. Experienced in requirement gathering, technical support, and collaboration with cross-functional teams to drive measurable business results through AI-powered solutions and scalable backend architectures.

Education

• Jain University August 2020 – June 2024
• Bachelor of Technology in Computer Science and Engineering - Artificial Intelligence CGPA: 8.15/10.0
• Core Technical Skills
• Programming & Coding : Python (PEP8)
• Backend Frameworks : Django, Flask, REST APIs, Microservices Architecture
• Databases & Big Data : MySQL, PostgreSQL, MongoDB, SQLite, Apache Hadoop, Apache Spark, PySpark, HDFS,
• Amazon S3
• Cloud & DevOps : Microsoft Azure, Docker, Jenkins, CI/CD Pipelines, Git, GitHub, YARN, MapReduce
• Analytics & AI : Machine Learning, Deep Learning, NLP, RAG, Big Data Processing, Business Analytics, TensorFlow,
• Scikit-Learn, XGBoost, LangChain, Hugging Face
• Enterprise Solutions : API Design, Web Architecture, Scalability, JWT Authentication, Production Systems Management
• Development Practices : Agile, SDLC, Version Control, OOP, Code Auditing, Technical Feasibility Assessment

Experience

Projects

• Customer Churn Prediction using Big Data & Machine Learning | Hadoop, PySpark, Scikit-Learn, TensorFlow
• Built end-to-end ML pipeline processing 500MB+ customer data using Apache Hadoop (HDFS) and PySpark for distributed computing, handling 1M+ records efficiently
• Implemented data preprocessing and feature engineering on distributed datasets using PySpark transformations including window functions, aggregations, and SQL operations
• Developed classification models achieving 87% accuracy using ensemble methods (Random Forest, Gradient Boosting) and deep learning (TensorFlow Neural Network)
• Engineered scalable data pipeline with HDFS storage, PySpark ETL processes, and automated model training workflows reducing processing time by 60%
• Deployed Dockerized Hadoop cluster with 3-node configuration (NameNode, DataNodes, YARN) for distributed processing and fault tolerance
• Performed comprehensive model evaluation using precision, recall, F1-score, and ROC-AUC metrics with cross-validation for robust performance assessment
• MultiPDF Chat App with RAG | Python, LangChain, GPT-3.5/GPT-4, Hugging Face, Vector DB
• Developed an intelligent document QA system using Retrieval-Augmented Generation (RAG) architecture with
• LangChain framework and OpenAI GPT models
• Implemented document parsing and text extraction pipelines for processing multiple PDF files with chunking strategies for optimal context retrieval
• Integrated vector embeddings using Hugging Face models for semantic search, enabling accurate document retrieval based on query similarity
• Leveraged open-source LLMs including GPT-3.5 and GPT-4 to create context-aware conversational AI with document grounding capabilities
• Credit Card Fraud Detection | Python, Scikit-Learn, Pandas, Imbalanced-Learn
• Built end-to-end machine learning pipeline for detecting fraudulent credit card transactions using ensemble methods and anomaly detection algorithms
• Addressed class imbalance using SMOTE and undersampling techniques, improving model recall for minority class by
• 35%
• Optimized model performance focusing on Area Under the Precision-Recall Curve (AUPRC) metric, achieving 0.89
• AUPRC score
• Evaluated multiple algorithms including Random Forest, XGBoost, and Logistic Regression with cross-validation and hyperparameter tuning
• Professional Development
• Technical Mentor | Coding Bootcamp
• Mentored junior developers in Python coding best practices, backend development fundamentals, and problem-solving techniques, conducting code reviews and maintaining knowledge base documentation to accelerate learner progress
• AWS Cloud Security Certification | Amazon Web Services
• Completed comprehensive cybersecurity bootcamp by Amazon Web Services, specializing in secure coding practices, cloud security architecture, and enterprise-grade security implementations for production systems
• Stakeholder Communication & Technical Leadership | Cappitall Want Network
• Led collaboration with business stakeholders through confident written and verbal communication, conducting requirement gathering sessions, technical feasibility assessments, and delivering presentations showcasing AI-powered
• FinTech solutions to cross-functional teams

Certifications

• MuleSoft Certified Developer - Level 1 | MuleSoft, Salesforce
• Completed industry-recognized certification validating expertise in API-led connectivity, integration design, and building scalable Mule applications for enterprise system integration
• Types of Cyber Security | Great Learning
• Completed professional certification covering core cybersecurity domains including network security, application security, secure coding practices, and threat mitigation strategies