PRANAY BAGARIA
AI EngineerAI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
AI Engineer with less than a year in Generative AI systems and RAG pipelines.
AI/ML engineering student with hands-on experience building production-grade Generative AI systems - RAG pipelines, multi-LLM applications, and generative content pipelines - using Python, FastAPI, LangGraph (LangChain ecosystem), ChromaDB/FAISS vector databases, and Docker. Developed a 10-architecture RAG evaluation platform (Llama 4 Scout, ChromaDB vector DB, LangGraph), an end-to-end generative AI audio pipeline (Llama-2 + MusicGen), and multi-LLM prompt-engineered automation workflows (Gemini, Claude, Groq) with cloud-ready Docker deployments. Brings practical LLM integration and prompt engineering experience from a remote startup internship and 8+ independently shipped AI projects.
BMS Institute of Technology and Management, Bengaluru
B.E. · Artificial Intelligence and Machine Learning
August 1, 2023 – June 30, 2027
VRB Capital LLC
AI Automations & IT Systems Intern
March 1, 2026 – April 30, 2026
India
Resume Intelligence System - Multi-LLM GenAI Application with RAG
June 1, 2026 – Present
Built a full-stack GenAI application (FastAPI REST backend + React frontend) processing unstructured resume data through a RAG pipeline - FAISS vector database retrieves context, then multi-LLM inference (Llama-3, Gemini, Claude) generates structured job-match scores, AI-written improvement suggestions, and personalized content. Integrated 3 LLM APIs (Gemini, Claude, Groq/Llama-3) with prompt-engineered chains for summarization, classification, and content generation; self-enriching feedback loop drives resume scores from baseline to >85/100 over iterations.
View ProjectRAG Studio - GenAI Platform: 10-Architecture RAG Evaluation System
June 1, 2026 – Present
Built a GenAI platform benchmarking 10 RAG architectures (Hybrid, Graph, Agentic, Corrective, Multimodal, RAG-Fusion, HyDE, Self-RAG, Structured, Multilingual) side-by-side on user-uploaded documents - powered by Llama 4 Scout 17B via Groq, ChromaDB (persistent vector database), BM25 sparse retrieval, and cross-encoder re-ranking; LangGraph (LangChain ecosystem) orchestrates agentic retrieval workflows with semantic query caching. Engineered 18 FastAPI endpoints for document ingestion, real-time SSE streaming, Compare Mode (all 10 architectures concurrently), and RAGAS-inspired 4-metric evaluation (Faithfulness, Relevance, Precision, Recall); prompt-engineered retrieval chains per architecture; Next.js 14 + React 18 frontend, Docker Compose deployment.
View ProjectCyber Wave Music Gen - Generative AI Audio Pipeline (LLM + MusicGen)
June 1, 2026 – Present
Architected an end-to-end generative AI content pipeline: user text prompt → Llama-2-7B (via Ollama, local open-weight LLM deployment) generates a descriptive music scene → MusicGen (Meta AudioCraft) produces audio output orchestrated via a custom LLM chaining layer with structured prompt engineering at each generation stage. Deployed a FastAPI backend serving async LLM + audio generation inference endpoints and a Next.js 14 React frontend for interactive music generation; used Ollama for local, cloud-independent open-weight LLM inference enabling on-device generation without external API costs.
View ProjectAI Support Ticket Automation - LLM Fine-tuning & NLP Pipeline
June 1, 2026 – Present
Fine-tuned DistilBERT (HuggingFace/PyTorch) on IT support ticket data achieving ~90%+ classification accuracy; built a fully automated NLP pipeline for ticket triage, priority assignment, and routing - replacing a manual classification workflow end-to-end. Added an LLM-powered post-processing layer with prompt-engineered templates for human-readable ticket summaries and suggested resolutions; surfaced classification accuracy, precision/recall, and model health metrics via a Streamlit analytics dashboard.
View ProjectCultural Fit Analysis
The candidate's portfolio showcases a strong passion for AI/ML, with diverse projects covering various aspects of Generative AI, from RAG systems to audio generation and NLP pipelines. This breadth of interest and continuous learning aligns well with an innovative and fast-paced AI engineering environment. The remote internship experience also indicates adaptability and self-management.
Soft Skills & Operational Fit
The candidate demonstrates strong initiative and a proactive approach through numerous personal projects. The internship experience shows an ability to integrate AI solutions into business workflows and manage IT infrastructure, indicating good operational fit. The focus on documentation and structured output suggests an organized approach to development.