Help & Supportcareers@gravityer.com
AI Engineer with 2+ years in Production AI/ML Systems & Data Science
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Recent Graduate and AI/ML Developer with 2.5+ years of production experience designing, deploying, and monitoring end-to-end AI/ML systems, spanning feature engineering, model training, deployment, and automated retraining across cloud. Proficient in Python and object-oriented design with strong CS fundamentals in data structures, algorithms, and complexity analysis. Hands-on expertise in ML model evaluation, explainable AI (XAI/SHAP), and responsible AI practices with applied experience in personalization systems, predictive customer analytics, and agentic AI orchestration for multi-step reasoning workflows.
Carleton University
Master of Engineering · Data Science, Analytics and Artificial Intelligence
August 1, 2025 – June 30, 2026
Anna University
Bachelor of Engineering · Computer Science and Engineering
N/A – June 30, 2022
DocumentPro
AI Engineer Intern
January 1, 2025 – June 1, 2025
India
Acies Global
Data Science Associate
August 1, 2023 – May 1, 2024
India
Postulate Info Tech Private Limited
Data Science Intern
January 1, 2023 – June 1, 2023
India
Suvidha Foundation
Machine Learning Intern & Volunteer
August 1, 2022 – January 1, 2023
India
Personalizing Medical Assistant with PEFT and RAG
June 24, 2026 – Present
Built a medical Q&A personalization system fine-tuning Mistral 7B with QloRA (PEFT) on 50K clinical records, achieving 0.68 ROUGE-L; deployed as Flask API with sub-500ms response time. Developed an end-to-end pipeline spanning PDF ingestion, document parsing, embedding generation, and Retrieval-Augmented Generation (RAG)-based response synthesis, with vector databases, grounding responses in structured patient records to reduce hallucinations by 40%+ vs. base model. Implemented responsible Al safeguards: PII-safe prompt design, output validation against retrieved source documents, & confidence-based fallback patterns to flag low-certainty responses, critical for regulated healthcare contexts. Evaluated model performance using semantic similarity metrics and ROUGE scores; tracked cost-per-query and latency across model variants to optimize the production deployment configuration. Built a React and Next.js frontend for the medical assistant, enabling real-time streaming responses via server-sent events, a conversational chat UI with source citation display, and API route integration with the Flask backend.
Osteoporosis Image Classification using SAM & Transfer Learning
June 24, 2026 – Present
Built end-to-end ML pipelines for X-ray image ingestion, preprocessing, segmentation & classification. Employed transfer learning with CNN architectures (DenseNet201, VGG19, ResNet50) for medical image classification; explored vision-language model (VLM) grounding techniques to align visual features with diagnostic text descriptions. Improved diagnostic classification accuracy from 54% to 78% using ROI segmentation (SAM) and transfer learning and also benchmarked multiple CNN architectures to optimize model robustness and performance.
IoT Anomaly Detection with XAI and TinyLlama
June 24, 2026 – Present
Built an end-to-end XAI anomaly detection system for smart home IoT security using the CASAS dataset, deploying a Random Forest classifier to identify abnormal sensor activity in real-time operational data streams. Integrated TinyLlama-1.1B (a compact, resource-efficient LLM suited for edge-constrained environments) to generate human-readable explanations of detected anomalies, bridging ML outputs and operational interpretability. Applied SHAP for feature-level explainability, providing transparent confidence scores and sensor-level attribution for each flagged event, critical for trust in real-world deployments. Deployed as a full-stack Streamlit application with date-based anomaly browsing, LLM explanation generation, and SHAP visualization, demonstrating end-to-end ownership from model to user-facing interface.
Cultural Fit Analysis
The candidate's project diversity (medical imaging, IoT, personalized Q&A) and experience across different types of organizations (internships, full-time, volunteer) indicate adaptability and a broad interest in applying AI. Their involvement in volunteer work and mentoring suggests a collaborative and community-oriented mindset, aligning well with a positive cultural fit. The target role of AI Engineer is well-aligned with their demonstrated skills and project focus.
Soft Skills & Operational Fit
The candidate demonstrates strong operational fit through their experience in CI/CD automation, model observability, and real-time monitoring. Their project descriptions highlight end-to-end ownership, collaboration with cross-functional stakeholders, and the ability to translate business requirements into scalable AI solutions. The mentoring experience also suggests leadership and communication skills.