Machine Learning Real-Time

Developed distributed microservices for a high-scale AI driven hiring platform, integrating gRPC APIs with GCP & Kubernetes, improving request efficiency and reducing latency by 65%. Built a real-time risk assessment system using Kafka & GCS, enabling event-driven compliance checks across 50 states, reducing processing time by 40%.

Designed and scaled high-throughput data pipelines with MongoDB & Redis, implementing indexing & caching strategies for low- latency retrieval of high-volume queries.

Deployed high-throughput ETL pipelines on GCP (Cloud Storage, BigQuery, Cloud Functions), processing 10M+ records daily with 99.8% uptime.

Optimized model inference APIs using Airflow and MLflow pipelines, improving serving throughput by 40% with automated deployment and monitoring.

Built real-time observability and alerting systems with Google Data Studio & Grafana, reducing incident response time by 25% and improving system monitoring.

GOKULA RANGA NAVEEN CHAPALA

585-***-**** ******@***.*** linkedin.com/in/gokul-naveen github.com/GokulNaveen2708 Rochester Institute of Technology, New York

Master of Science in Computer Science

Aug. 2023 – Aug. 2025

EDUCATION

EXPERIENCE

PROJECTS AND RESEARCH

SKILLS

Scalable Data Systems for Kaggle Analytics

Designed a hybrid PostgreSQL + MongoDB data pipeline to process 126M records, optimizing query execution speed by 98% through advanced indexing and performance tuning.

Implemented itemset mining (Apriori) on 1.4M submissions, extracting tag correlations and high achiever patterns, and developed automated Tableau dashboards for real-time competition analytics. Languages : Python, Java, C++, JavaScript, SQL, C#. Cloud & DevOps: AWS (EC2, S3, Lambda, DynamoDB), GCP, Azure, Kubernetes, Docker, Terraform, Git, Jenkins, CI/CD. Software Development: FastAPI, Flask, gRPC, GraphQL, Microservices, Distributed Systems, REST APIs. Machine Learning & AI: TensorFlow, PyTorch, Hugging Face, BERT, GPT, RAG, MLflow. Data Visualization & Monitoring: Tableau, Google Data Studio, Elasticsearch, Logstash, Grafana. Jul. 2021 – Jul. 2023

Distributed Trust-Aware Federated Learning System

Designed a trust-based federated learning system in Flower, boosting model robustness by 35% under adversarial attacks. Reduced poisoned update impact by 60% using dynamic trust scoring and selective client aggregation with near-zero overhead. Achieved 87% model accuracy (vs. 80% baseline) in non-IID settings by filtering malicious clients based on deviation metrics. Coursework: Computational Problem Solving, Advanced Object-Oriented Concepts, Machine Learning, Artificial Intelligence, Analysis of Algorithms, Intro to Big Data, Intelligent Security Systems, Distributed Systems, Big Data Analytics, Database Systems Software Engineer Accenture

Hyderabad, TG

Software Engineer Intern Zoho

Chennai, TN

Jan. 2021 – Jul. 2021

Developed and optimized API endpoints for task management using Spring Boot & PostgreSQL, improving query efficiency and reducing response time by 15%.

Implemented gRPC-based microservices and optimized database performance with read replicas and connection pooling, enhancing scalability and high-traffic data retrieval.

Integrated Redis caching and API rate limiting, ensuring efficient request handling, and improved CI/CD workflows with Kubernetes

& Terraform, automating deployments and scaling.

Graduate Course Assistant RIT

Rochester, NY

Jan. 2025 – May. 2025

Designed and implemented cryptographic algorithms (RSA, ECC, Zero-Knowledge Proofs) for course demonstrations, developing secure coding exercises and backend implementations to reinforce real-world applications. Developed hands-on labs and automated grading scripts for cryptographic protocols, integrating secure authentication mechanisms

(JWT, OAuth2, HMAC) to teach best practices in system security and backend development. RAFT Consensus Algorithm

Implemented the RAFT consensus algorithm in Java, enabling scalable log replication and leader election across 256 nodes. Deployed a fault-tolerant RAFT cluster using Docker & Docker Compose, automating the launch of nodes and clients for distributed coordination.

Scalable Token Level Watermarking in LLMs

Designed and optimized token sampling and detection logic with <100ms overhead on 512-token sequences, enabling real-time inference compatibility.

Delivered high-accuracy watermark detection (<2% false positives) across LLM outputs without degrading generation quality or performance.

Contact this candidate