Help & Supportcareers@gravityer.com
AI Engineer with 1+ years in LLMs, RAG, and Cloud Deployment
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 Engineer with research and hands-on engineering experience in large language models. First-authored a research paper on numerical robustness in Transformer inference (TCS-Intel) and fine-tuned LLaMA-3.2-3B with QLORA and GGUF quantization for edge deployment. Skilled in building RAG pipelines, autonomous AI agents (LangGraph), and Python backends, and deploying scalable systems on AWS with Docker.
Indian Institute of Information Technology Lucknow
M.Sc Artificial Intelligence and Machine Learning (AI & ML) · Artificial Intelligence and Machine Learning
August 1, 2024 – June 30, 2026
Marwari College Ranchi
B.Sc Mathematics (Honours) · Mathematics
August 1, 2020 – June 30, 2023
TCS Research (Intel Collaboration)
Research Intern
December 1, 2025 – April 30, 2026
India
INAI Worlds
AI Engineer Intern
July 1, 2025 – November 30, 2025
India
Custom AI Coding Assistant LLAMA-3.2-3B Fine-Tuned
January 1, 2024 – Present
Instruction-tuned LLaMA-3.2 (3B) via 4-bit QLORA (NF4) on 19,400+ Python samples achieving 77.9% token accuracy; merged LORA weights and applied Q4_K_M GGUF quantization to compress the model by 69% (6GB→ 1.88GB) for edge deployment. Deployed via Ollama for secure offline real-time inference on consumer hardware (4GB+ VRAM).
View ProjectAutonomous Regulatory Compliance Agent (ARCA)
January 1, 2024 – Present
Architected a stateful 7-agent LangGraph pipeline to automate end-to-end banking compliance, parsing dense regulatory circulars (PyMuPDF + OCR) and extracting actionable tasks via Chain-of-Thought reasoning. Engineered semantic RAG routing by indexing bank department profiles in ChromaDB, autonomously distributing compliance mandates across 13 departments with a secondary LLM verification loop ensuring 98% accuracy. Developed a full-stack dashboard using React, Node.js, and Express connected to a Prisma/PostgreSQL database, featuring real-time WebSocket event streaming for live compliance tracking and automated JIRA/SMTP escalation.
View ProjectAcademic Sloth: AI-Powered Research Assistant
January 1, 2024 – Present
Architected an Agentic RAG pipeline using LangGraph to orchestrate a multi-agent system; built a Router Agent that classifies user intent to trigger specialized agents (Deep-Dive, Compare, Conversational) over academic PDFs. Developed a hybrid retrieval system (BM25 + vector search) with Cross-Encoder re-ranking and an automated Grounding Guard agent that fact-checks LLM outputs against source chunks to suppress hallucinations.
View ProjectCultural Fit Analysis
The candidate's academic background in AI/ML and Mathematics, coupled with diverse personal and academic projects, demonstrates a strong passion for the field and a continuous learning mindset. The involvement in hackathons and competitive programming indicates a drive for excellence and a competitive spirit. The range of technologies and project types suggests adaptability and a willingness to explore different domains within AI, aligning well with an innovative and fast-paced environment.
Soft Skills & Operational Fit
The candidate's project descriptions and experience indicate strong problem-solving skills, a proactive approach to learning and applying advanced AI concepts, and the ability to work on complex, multi-faceted projects. The leadership role as Placement Coordinator and GenAI wing Coordinator suggests good organizational and team collaboration potential. The detailed descriptions of technical challenges and solutions imply a structured and analytical thought process.