Data Engineer - ML Engineer Intern with ETL, ML Ops, and RAG

Vaibhav Singh Chandel

626-***-**** **********@*****.*** linkedin.com/in/vybhv github.com/stonedseeker Experience

Data Engineering Intern August 2024 – Present

Motorola Solutions Bangalore, Karnataka

• Engineered ETL pipelines in Python and SQL to extract, transform, and load log data from distributed microservices, enabling faster and reliable insights.

• Streamlined data processing workflows for cleaning, aggregation, and KPI tracking, boosting debugging efficiency and reducing verification time by 70%.

• Built interactive dashboards for performance monitoring and anomaly detection, cutting manual effort by 60% and accelerating root-cause analysis.

Machine Learning Intern March 2023 – June 2023

Group Cyber ID Technology Bangalore, Karnataka

• Enhanced surveillance efficiency by deploying an open-source ANPR system with YOLOv8, PyTorch, and EasyOCR that facilitates real-time license plate detection and data extraction from CCTV footage.

• Implemented an end-to-end solution by deploying the model on Google Cloud Platform, establishing a streamlined pipeline into the existing infrastructure, ensuring scalability and efficient model updates.

• Employed image processing techniques with OpenCV and dataset annotation via Roboflow, resulting in a 25% improvement in the model’s resilience to handle challenging and distorted data. Projects

Pactum AI Live GitHub PyTorch, Transformers, PEFT, TRL, FastAPI October 2025

• Designed and deployed autonomous multi-agent system with specialized agents (analyzer, negotiator, orchestrator) for contract analysis, implementing agent orchestration and task coordination workflows using prompt engineering and Chain-of-Thought reasoning.

• Formulated end-to-end LLM pipeline including LoRA fine-tuning on GPT-2/Phi-2 models (84K+ contracts), PPO-based RLHF with custom reward modeling for alignment, and 4-bit quantization achieving 75% memory reduction for efficient deployment.

• Deployed production agent system with RAG pipeline, FastAPI endpoints, and comprehensive evaluation benchmarks tracking clause extraction accuracy, risk assessment quality, and model performance metrics in real-world conditions.

Voyage AI GitHub Pinecone, Neo4j, OpenAI GPT-4, LangChain, Python October 2025

• Architected knowledge-grounded agent system using RAG with Pinecone vector search and Neo4j graph database

(360 entities, 700+ relationships), implementing multi-hop reasoning and context-driven prompting for intelligent travel recommendations.

• Developed advanced prompt engineering strategies including semantic query expansion (30% better recall), two-layer relevance detection eliminating hallucinations, and LRU embedding cache reducing API costs 60%.

• Built production deployment with error handling, hyperparameter optimization, and evaluation framework measuring citation accuracy (95%), response quality, and latency metrics across diverse user queries. Assemblage GitHub HDBSCAN, Sentence-Transformers, SQLite, OpenAI GPT-4o October 2025

• Engineered complete ML pipeline from data acquisition (8K+ app reviews) through preprocessing, feature engineering (sentence embeddings), unsupervised clustering (HDBSCAN), to LLM-powered insight generation with prompt optimization strategies.

• Designed scalable production system with GPU acceleration, embedding cache, and SQLite persistence, implementing validation strategies and performance benchmarks achieving 90%+ deduplication accuracy while reducing costs 1200x vs LLM-only approaches.

Toggle Live GitHub ChromaDB, OpenAI GPT-4o, Serper.dev, Firestore October 2025

• Modeled hybrid RAG system with parallel document and web search retrieval, implementing intelligent ranking

(similarity >0.7) and session isolation achieving 60% API cost reduction across 10+ concurrent users. Education

REVA University, Bangalore, Karnataka Nov 2021 – July 2025 CGPA: 8.99 B.Tech in Artificial Intelligence & Data Science

Contact this candidate