Help & Supportcareers@gravityer.com
AI Engineer with 1+ years in Machine Learning, LLM, and Computer Vision.
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Final-year IIT Delhi student targeting full-time AI Engineer roles from June 2026. Strong foundation in machine learning, LLM fine-tuning, retrieval-augmented generation, computer vision, and applied data workflows. Experienced in dataset creation, model training/evaluation, inference pipelines, APIs, vector search, and knowledge graphs using Python, PyTorch, TensorFlow, FastAPI, FAISS, Qdrant, and Neo4j.
Indian Institute of Technology Delhi
B.Tech · Electrical Engineering, Power and Automation
N/A – June 30, 2026
Chunchreek Ventures Pvt. Ltd.
AI/ML Intern
August 1, 2025 – May 1, 2026
India
iKites Technologies
Research Intern
May 1, 2025 – July 1, 2025
India
Domain-Adapted Qwen 8B for Electrical Engineering and Embedded Systems
April 1, 2026 – Present
Fine-tuned a distilled Qwen 8B model using QLoRA 4-bit training for domain adaptation in electrical engineering and embedded systems. Built a 9K instruction-response dataset from circuits, operating systems, and computer architecture textbooks. Designed a teacher-student distillation pipeline where a stronger LLM generated step-by-step reasoning traces for technical problem solving. Performed SFT and reasoning fine-tuning; optimized single-GPU training with paged optimizers, gradient checkpointing, low-bit quantization, and memory-efficient batching.
AI-Powered PDF Q&A Platform with Hybrid RAG (Vector + Knowledge Graph)
December 1, 2025 – January 1, 2026
Built a real-time AI-powered PDF Q&A platform using React, FastAPI, PostgreSQL, Neo4j, and OpenAI GPT-40. Developed a Hybrid RAG pipeline combining vector embeddings and Knowledge Graph retrieval for improved contextual accuracy. Implemented a real-time OCR-based PDF ingestion pipeline with parallel chunk extraction, schema validation, and cross-page context merging. Designed a streaming conversational chat system with JWT authentication, session memory, and markdown-rendered responses. Architected modular backend services for parsing, embedding, KG construction, retrieval, and scalable multi-user document processing.
Circuit Extraction from PDF Datasheets using YOLO
December 1, 2024 – January 1, 2025
Developed a computer vision pipeline to detect and extract electrical circuit diagrams from heterogeneous PDF datasheets. Created annotations using Microsoft VoTT, trained a YOLO detector, and converted detected regions into retrievable context. Built a React.js retrieval assistant combining YOLO outputs with the Gemini API; achieved 92% detection accuracy.
Physics-Informed Neural Networks for Structural Analysis
May 1, 2024 – June 1, 2024
Built Physics-Informed Neural Networks in TensorFlow for harmonic oscillator and Poisson equation solvers using automatic differentiation and domain-specific losses. Achieved 98% and 92% accuracy on the respective equation-solving tasks.
Cultural Fit Analysis
The candidate's academic background from IIT Delhi and involvement in diverse AI projects (medical, electrical engineering, general Q&A) suggest a strong drive for learning and innovation. Their experience in an early-stage startup environment (Chunchreek Ventures) indicates adaptability and a proactive attitude, which aligns well with dynamic, fast-paced environments. The breadth of technologies and problem domains tackled reflects intellectual curiosity and a willingness to explore new challenges.
Soft Skills & Operational Fit
The candidate demonstrates strong problem-solving skills through complex project implementations. Their involvement in product strategy and investor discussions at Chunchreek Ventures suggests an understanding of business context. Collaboration with professors and participation in team projects indicate good teamwork potential. The academic focus on electrical engineering combined with AI projects shows a multidisciplinary approach.