Help & Supportcareers@gravityer.com
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 Engineer with less than a year in LangGraph & RAG.
AI/ML Engineer with expertise in LangGraph, RAG, and Agent-based systems. Currently gaining practical experience as an Artificial Intelligence Applications Development Intern, specializing in document processing workflows, vision extraction pipelines, and robust output validation. Proficient in building full-stack RAG applications, scalable multi-agent research engines, and generative AI dialogue summarizers, leveraging skills across Python, FastAPI, Docker, and various ML/DL frameworks. Seeking to apply strong problem-solving and implementation skills in an innovative AI/ML environment.
VIT Bhopal University
B.Tech · Computer Science and Engineering
August 1, 2023 – June 30, 2027
AmberFlux EdgeAI Private Limited
Artificial Intelligence Applications Development Intern
May 1, 2026 – Present
Hyderābād, Telangana, India
Argus - Scalable Multi-Agent Research Engine
February 1, 2026 – February 1, 2026
Built a LangGraph supervisor-based multi-agent pipeline with 5 specialist agents (planner, researcher, critic, writer, supervisor) using LLM-driven Command(goto) routing to autonomously synthesize cited markdown reports - reducing manual research time from hours to 30-90 seconds. Designed async submit → poll → fetch job flow with SQLite persistence and LangGraph checkpointing for failure recovery; traced every LLM call end-to-end in LangSmith for observability. Containerized with Docker, deployed on Render with a health-check endpoint, and integrated Tavily web search, ArXiv, and Wikipedia as agent tools.
View ProjectDoCopilot (RAG Document Q&A)
December 1, 2025 – December 1, 2025
Built a production full-stack RAG app: upload PDFs/TXT, index with Qdrant hybrid search (BM25 + dense vectors), and retrieve cited answers; deployed with a FastAPI backend and Next.js frontend. Improved retrieval quality via RRF fusion + cross-encoder reranking; added guardrails (prompt-injection detection, PII redaction, source-grounding checks) to harden the pipeline against adversarial inputs. Ran a 40-query LLM-as-Judge ablation study reporting 89.2% avg correctness, 90.5% relevance, 100% source rate, and 2.86s avg latency - demonstrating measurable retrieval quality over baseline RAG.
View ProjectFLAN-T5 Dialogue Summarizer (GenAI)
October 1, 2025 – October 1, 2025
Deployed an interactive Gradio app on Hugging Face Spaces with configurable decoding (beam search, max length); published the fine-tuned model with a full model card and reproducible evaluation suite (ROUGE, BERTScore, METEOR, BLEU). Fine-tuned FLAN-T5-base with LoRA on SAMSum (14.7K dialogues), achieving 49.01 ROUGE-1, 72.25 BERTScore F1, and 42.51 METEOR. Implemented parameter-efficient training (LoRA r=16, α=32), updating only 2% of parameters with FP16 mixed precision - matching full fine-tuning ROUGE-1 at a fraction of the compute cost.
Deep Learning Specialization
deeplearning.ai / Coursera
January 1, 2025 – Present
Applied Machine Learning in Python
Coursera
January 1, 2025 – Present
LangChain for LLM Application Development
deeplearning.ai
January 1, 2025 – Present
Cultural Fit Analysis
The candidate's personal projects showcase a proactive and self-driven learning attitude, exploring diverse areas within AI/ML from RAG to multi-agent systems and fine-tuning. The detailed descriptions and quantitative results in projects like DoCopilot and FLAN-T5 demonstrate a commitment to quality and measurable impact. The current internship further solidifies their practical application of AI in an enterprise setting, indicating a good fit for a role requiring hands-on development and problem-solving.
Soft Skills & Operational Fit
The candidate demonstrates strong problem-solving skills through the implementation of complex AI systems and optimization techniques. The detailed project descriptions suggest a methodical approach to development and a focus on measurable outcomes. The use of observability tools like LangSmith indicates an understanding of operational best practices for AI applications.