Help & Supportcareers@gravityer.com
AI Engineer with less than a year in RAG pipelines, multi-agent systems, and LLM evaluation framewor
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/LLM Engineer fresher with hands-on experience building RAG pipelines, multi-agent systems, and LLM evaluation frameworks using LangChain, LangGraph, and LlamaIndex. Skilled in designing modular AI architectures with FastAPI backends and Streamlit UIs, deploying production-grade Generative AI products. Strong communicator with proven ability to collaborate cross-functionally and deliver results under pressure.
Acropolis Institute of Technology and Research, Indore
Bachelor of Science · Information Technology
August 1, 2021 – June 30, 2025
The St Peters Higher Secondary School, Indore
Class 10th & 12th
June 1, 2017 – May 31, 2020
RAG Pipeline with RAGAS Evaluation
January 1, 2026 – June 1, 2026
Architected an end-to-end RAG pipeline ingesting 1 PDF document with semantic chunking, embedding generation, and vector retrieval using LlamaIndex, improving retrieval accuracy by an estimated 20% over keyword-based search. Integrated RAGAS evaluation framework measuring faithfulness, answer relevancy, context precision, and context recall across a 30-sample evaluation dataset, achieving 85%+ average scores across all metrics. Designed a modular src/ architecture separating ingestion, chunking, embeddings, retrieval, and evaluation modules, enabling reproducible benchmarking and collaborative extensibility for future team contributions. Containerized the full pipeline with Docker for environment-independent execution across development and production environments.
View ProjectMulti-Agent AI Trip Planner
January 1, 2026 – June 1, 2026
Built a multi-agent travel planning system using LangGraph with 4 specialized agents, conditional routing, and shared state management, reducing itinerary generation time by ~40% vs. single-agent approaches. Integrated 5 LLM-powered tools for dynamic itinerary generation, search, and preference-aware travel recommendations via prompt engineering, improving recommendation relevance through iterative hyperparameter tuning of temperature and top-p settings. Developed a FastAPI backend exposing agentic API endpoints and a Streamlit frontend delivering an interactive conversational interface with sub-2-second response times through performance optimization of the inference pipeline.
View ProjectLocal AI Assistant with Offline LLM
January 1, 2026 – June 1, 2026
Designing a privacy-first AI assistant running entirely on local LLMs via Ollama, eliminating dependency on external API services and enabling offline model deployment for sensitive enterprise use cases. Implementing tool-calling workflows and conversational memory using LangGraph for stateful multi-turn interactions with cloud computing-independent architecture. Integrating MCP (Model Context Protocol) for standardized agent-tool communication and extensible tool registration, with attention to data preprocessing pipelines for diverse local document formats.
View ProjectAWS Certified AI Practitioner (AIF-C01)
AWS
January 1, 2025 – Present
DBMS Certification
NPTEL
January 1, 2025 – Present
Cultural Fit Analysis
The candidate's project diversity, focusing on privacy-first AI, multi-agent systems, and RAG pipelines, aligns well with an innovative and problem-solving culture. The AWS certification and active participation in a cloud enthusiasts club indicate a proactive learning attitude and engagement with the broader tech community. However, the lack of professional work experience might indicate a need for mentorship in a fast-paced corporate environment.
Soft Skills & Operational Fit
The candidate's involvement in leading the AWS Cloud Enthusiasts Club and securing a top 10 position in a college hackathon suggests strong communication, leadership, and teamwork skills. The ability to deliver cloud-based solutions under pressure indicates good operational fit and stress handling. However, without specific psychometric test results, a deeper assessment of work attitude and team collaboration is limited.