AI Engineer with less than a year in RAG systems & LLMs
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
GenAI Engineer (Fresher) skilled in RAG systems, LLMs, LangChain, FAISS, FastAPI, and cloud deployment (AWS EC2, Docker). Built an end-to-end PDF Q&A system powered by Llama 3.3 70B achieving 93% faithfulness on RAGAS evaluation, monitored with LangSmith for LLM tracing and latency tracking, and an ML-backed insurance prediction API deployed with Docker on AWS. Strong foundation in Python, Hugging Face Transformers, vector databases, and REST API design. Looking to apply practical GenAI and MLOps skills in an internship or entry-level AI role.
D. Y. Patil Agriculture and Technical University, Talsande
B.Tech · CSE (AI & ML)
August 1, 2024 – June 30, 2027
Sant Gajanan Maharaj Rural Polytechnic, Mahagaon
Diploma · Computer Engineering
August 1, 2021 – June 30, 2024
Wayspire
Artificial Intelligence Intern
July 1, 2025 – September 1, 2025
India
RAG-Powered PDF Question Answering Chatbot
January 1, 2026 – May 1, 2026
Built an end-to-end document Q&A system enabling users to upload PDFs and ask natural language questions with context-aware answers powered by Llama 3.3 70B via Groq. Implemented document chunking, Hugging Face sentence-transformers embeddings, and FAISS vector store for semantic retrieval; cut ingestion time from ~90s to under 3s via FAISS index persistence. Achieved 93% average faithfulness score across evaluation questions using the RAGAS framework, confirming near-zero hallucination rate. Integrated LangSmith for end-to-end LLM pipeline observability traced all LLM calls (inputs, outputs, intermediate steps), monitored token usage and latency per query, and used evaluation runs to debug and improve retrieval quality. Deployed FastAPI backend on AWS EC2 via Docker with API secrets secured in AWS SSM Parameter Store; built interactive Streamlit UI reducing document search time significantly.
View ProjectInsurance Premium Prediction System
January 1, 2026 – June 1, 2026
Developed a RESTful backend using FastAPI to process user demographic and lifestyle data for ML-based insurance premium prediction; integrated Pydantic for robust schema validation and error handling. Containerized the full application with Docker and deployed on AWS EC2, ensuring consistent and reproducible deployments. Integrated trained ML model inference into the FastAPI backend and built a Streamlit frontend for interactive user access.
Python Programming Certificate
IBM Developer Skills Network
June 1, 2026 – Present
Artificial Intelligence Internship Certificate
Wayspire
June 1, 2026 – Present
Git Training Certificate
Simplilearn
June 1, 2026 – Present
Data Structures & Problem Solving Certificate
GeeksforGeeks
June 1, 2026 – Present
Cultural Fit Analysis
The candidate's projects show a strong interest and initiative in cutting-edge AI/ML areas, particularly GenAI. The diversity of projects (RAG chatbot, insurance prediction, fake news detection) indicates a broad curiosity and willingness to tackle different problem domains. The use of modern tools and frameworks aligns well with an innovative and fast-paced technical culture. The candidate is currently pursuing a B.Tech in CSE (AI & ML), which directly aligns with the target role. However, the lack of team-based project experience or open-source contributions makes it difficult to fully assess collaboration and broader cultural fit beyond individual technical drive.
Soft Skills & Operational Fit
The candidate demonstrates strong problem-solving skills, particularly in identifying and resolving dataset bias in the Fake News Detection system. Their project descriptions indicate a methodical approach to development, including evaluation and observability. The use of Docker and AWS EC2 suggests an understanding of operational best practices for deployment. However, as an entry-level candidate, further development in collaborative project management and advanced system design thinking would be beneficial.