AI Engineer with less than a year in LLM & Data Science
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/ML Intern at Docu3C Technologies, specializing in AI/LLM-based applications for financial document intelligence. Possessing a strong foundation in building end-to-end machine learning and LLM pipelines, I am proficient in leveraging advanced techniques like RAG, knowledge graphs, and multimodal emotion recognition. My expertise spans scalable backend development with FastAPI, interactive UI design, and cloud deployments on Azure. I am committed to delivering high-quality, optimized AI systems through rigorous testing and iterative refinement.
VIT Bhopal University
B. Tech · CSE (Core)
August 1, 2023 – Present
Docu3C Technologies
AI/ML Intern
October 1, 2025 – Present
India
WESAD Multimodal Emotion Recognition
October 1, 2025 – Present
Implemented and compared ML models (LDA, Random Forest, Decision Tree, AdaBoost, KNN) for emotion classification. Processed multimodal physiological signals (ECG, EDA, EMG, Respiration, Temperature, BVP, ACC). Designed experiments to evaluate sensor modality combinations (chest vs wrist vs combined). Built pipelines for cross-validation, accuracy, and F1-score evaluation. Generated detailed reports identifying optimal modality configurations.
View ProjectHybrid Legal RAG: Sparse-Dense-Graph Retrieval System
October 1, 2025 – Present
Engineered a production-grade retrieval-augmented system combining BM25 (FTSS), dense embeddings (MPNet + FAISS), and graph-based ranking (PageRank). Implemented Reciprocal Rank Fusion (RRF) to unify multiple retrieval modalities for improved ranking quality. Processed and indexed 26K+ documents, enabling high-precision semantic and keyword-based search. Achieved 82.4% Top-1 accuracy and NDCG@10 of 0.89, outperforming single-modality baselines. Built a FastAPI backend supporting search, conversational querying, and structured summarization. Conducted ablation studies to validate robustness under high-noise conditions. Containerized deployment using Docker with memory-efficient optimizations.
View ProjectPolarBrief AI – Legal Argument Analyzer
October 1, 2025 – Present
Developed an AI-powered document analysis tool to extract, classify, and score arguments from unstructured PDFs. Integrated LLMs via LangChain for summarization, heading generation, and structured scoring. Implemented argument polarity classification (Pro/Con/Neutral) with weighted ranking using TF-IDF + LLM outputs. Designed a hybrid OCR + text extraction pipeline using Tesseract and PDF parsers. Generated structured outputs in JSON/PDF formats with citation-aware summaries. Built an interactive Streamlit interface for visualization and report generation.
View ProjectKnowledge Graph Pipeline for Document Intelligence
October 1, 2025 – Present
Built an end-to-end knowledge graph extraction pipeline from unstructured documents using NLP and LLMs. Implemented multi-stage processing: OCR, cleaning, chunking, entity-relation extraction, and validation. Designed deterministic ID mapping and cross-chunk entity resolution using fuzzy matching. Normalized entities and relations to construct a clean, canonical knowledge graph. Visualized graphs using NetworkX and PyVis for interactive exploration. Exported structured graph data for downstream analytics.
View ProjectAzure Data Fundamentals
Microsoft
June 1, 2025 – Present
Introduction to Machine Learning
NPTEL
May 1, 2025 – Present
Prompt Design in Vertex AI
Google Cloud
July 1, 2024 – Present
Introduction to Generative AI
Google Cloud
April 1, 2024 – Present
Introduction to Large Language Models
Google Cloud
April 1, 2024 – Present
Cultural Fit Analysis
The candidate's diverse project portfolio, ranging from multimodal emotion recognition to legal RAG systems and knowledge graphs, indicates adaptability and a broad interest in AI applications. Their involvement in organizing technical workshops and participating in team sports suggests a collaborative mindset and engagement beyond academics, which are positive indicators for cultural fit within a dynamic engineering team. The certifications in Azure and Google Cloud also show a commitment to continuous learning and staying current with industry trends.
Soft Skills & Operational Fit
The candidate demonstrates strong problem-solving skills through LeetCode achievements and project complexities. Their experience as an Event Lead suggests organizational and team collaboration abilities. The detailed project descriptions indicate good communication of technical concepts. The candidate's current internship and project work align well with an AI Engineer role, showing a proactive and hands-on approach to learning and application.