
AI Engineer with 1+ years in AI/ML & Data Engineering
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 and Data Expert with over 1 year of experience designing structured AI workflows, building data processing pipelines, and developing backend microservices. Proficient in Python, FastAPI, and LangChain/LangGraph to build stateful multi-agent systems and full-stack RAG architectures. Adept at handling core machine learning algorithms, engineering high-dimensional vector embeddings, and creating interactive user interfaces for AI-driven solutions.
L.D. Engineering College
Bachelor of Engineering · Electrical Engineering
August 1, 2021 – June 30, 2025
Semantic Movie Recommender System
January 1, 2025 – June 1, 2026
Developed an end-to-end content-based recommendation engine utilizing deep learning-based sentence embeddings. Engineered 384-dimensional vector embeddings and calculated cosine similarity matrices for real-time recommendation mapping. Integrated the OMDb REST API to dynamically fetch and display metadata, IMDb ratings, and summaries. Built a modular FastAPI backend serving enriched JSON results to an interactive Streamlit user interface.
Advanced AI Chatbot (LangGraph Workflow)
January 1, 2025 – June 1, 2026
Designed and built a stateful AI application utilizing a graph-based framework to execute multi-step conversational problem-solving. Developed automated, structured workflows using LangGraph to manage complex conversation states and agent routing. Implemented custom logical branching and memory retention mechanisms to ensure accurate, multi-turn NLP context. Integrated state-of-the-art Large Language Models (LLMs) via API to handle diverse user queries and dynamic tasks. Leveraged Python and data structures to parse, clean, and format system inputs and agent responses efficiently.
AI YouTube Video Assistant (RAG-based Chatbot)
January 1, 2025 – June 1, 2026
Built a full-stack Retrieval-Augmented Generation (RAG) application enabling real-time semantic search and interaction with video content. Implemented LangChain and FAISS for document parsing, transcript extraction, and vector-based semantic indexing. Developed a high-performance FastAPI backend using Uvicorn to orchestrate REST API endpoints. Integrated LLMs (Gemini/Claude) via OpenRouter with HuggingFace embeddings for accurate contextual analysis. Designed a clean dual-pane UI with Streamlit to deliver an intuitive user experience for video viewing and chat.
Credit Card Fraud Detection (Machine Learning)
January 1, 2025 – June 1, 2026
Designed and optimized predictive machine learning models to detect fraudulent transactions in highly imbalanced financial datasets. Conducted data preprocessing, exploratory data analysis (EDA), and strategic feature engineering using Pandas and NumPy. Addressed extreme class imbalance by applying advanced resampling techniques and strategic model hyperparameter tuning. Trained and evaluated multiple classifiers (Logistic Regression, Random Forest, XGBoost), achieving optimal performance with a tuned XGBoost model.
Cultural Fit Analysis
The candidate's projects are diverse within the AI/ML domain, covering recommendation systems, chatbots, and fraud detection. This breadth, combined with the stated eagerness to learn and adopt new MLOps tools, suggests a good cultural fit for an innovative and evolving AI team. The focus on personal projects indicates self-motivation and initiative, which are positive cultural attributes. However, the lack of professional experience limits the assessment of collaboration in a team setting.
Soft Skills & Operational Fit
The candidate highlights problem-solving, communication, teamwork, and adaptability as soft skills. The project descriptions show an ability to break down complex technical problems and integrate various tools, suggesting a good operational fit for dynamic AI development environments. However, without direct interaction or peer feedback, these are self-reported and cannot be fully validated.