"Strategic automation is not about replacing human judgment—it's about amplifying engineering excellence through intelligent systems that scale quality, reliability, and innovation across the entire development lifecycle."

My approach combines technical depth with business acumen, focusing on:

• Proactive Quality Engineering: Building ML-validated frameworks that prevent issues before production
• Data-Driven Decision Making: Leveraging metrics and observability to guide strategic initiatives
• Cross-Functional Collaboration: Bridging gaps between QA, DevOps, and data science teams
• Continuous Innovation: Implementing cutting-edge AI/ML techniques to solve complex business challenges

graph TB
    A[Strategic Vision:<br/>AI-Powered Quality Engineering] --> B[Technical Leadership]
    A --> C[Process Innovation]
    
    B --> D[ML-Integrated CI/CD]
    B --> E[Automated Testing Frameworks]
    B --> F[Data Validation Pipelines]
    
    C --> G[30% Faster Deployments]
    C --> H[40% Fewer Data Incidents]
    C --> I[25% Quality Improvement]
    
    D --> J[Business Impact:<br/>Zero Critical Defects]
    E --> J
    F --> J
    G --> J
    H --> J
    I --> J
    
    style A fill:#1e3a8a,stroke:#3b82f6,stroke-width:3px,color:#fff
    style J fill:#059669,stroke:#10b981,stroke-width:3px,color:#fff
    style D fill:#7c3aed,stroke:#a78bfa,stroke-width:2px,color:#fff
    style E fill:#7c3aed,stroke:#a78bfa,stroke-width:2px,color:#fff
    style F fill:#7c3aed,stroke:#a78bfa,stroke-width:2px,color:#fff

timeline
    title Professional Growth Journey
    section Education Foundation
    2020-2024 : B.Tech Computer Science
              : IoT Specialization
              : Parul University
    section Industry Experience
    Dec 2023 : Data Science Intern
             : SlashMark
             : 96.3% Model Accuracy
    May 2024 : AI Automation Engineer
             : Jignect
             : 40% Incident Reduction
    section Advanced Education
    Aug 2025 : M.S. AI Systems
             : University of Florida
             : Strategic Leadership Focus

┌─────────────────────────┐
│ Test Suites: 50+        │
│ Critical Defects: 0     │
│ Quality Gain: +25%      │
│ Deployment Speed: +30%  │
│ Data Incidents: -40%    │
│ Failure Rate: -20%      │
└─────────────────────────┘

Executive Summary: Developed ML solution predicting workforce attrition with 92% accuracy, enabling proactive retention strategies

Technical Approach:

• Analyzed 50K+ employee records using logistic regression
• Engineered features improving F1-score by 18%
• Implemented fairness-aware evaluation (demographic parity, equalized odds)

Business Impact: Enabled HR to identify at-risk employees early, reducing recruitment costs and knowledge loss

Challenge: Manual processing of 100K+ survey responses creating bottlenecks

Solution:

• Deployed generative AI summarization pipeline with 95% accuracy
• Accelerated preprocessing workflows by 45%
• Implemented hallucination detection for quality assurance

ROI: Freed analysts from repetitive tasks, enabling focus on strategic insights

Technical Leadership:

• Optimized IoT data processing with JSON-MQTT protocol
• Reduced AI inference cost on edge devices by 30%
• Implemented model quantization shrinking size by 15%

Strategic Value: Demonstrated scalable edge AI architecture for real-time urban systems

Project Scale: 9 commits, comprehensive documentation, production-ready deployment

Technical Highlights:

• YOLOv8-based real-time emotion detection system
• Implemented SHAP explainability for model transparency
• Dockerized deployment with Prometheus/Grafana monitoring

Achievements:

• 99.5% mAP@0.5 detection accuracy
• 0.3ms inference time (1000x faster than target)
• Full monitoring stack with Grafana dashboards

mindmap
  root((AI/ML Engineering<br/>Expertise))
    Machine Learning
      Supervised Learning
      Unsupervised Learning
      Deep Learning
      Model Optimization
    AI Frameworks
      TensorFlow
      PyTorch
      Keras
      Hugging Face
      LangChain
    Data Engineering
      Apache Spark
      Great Expectations
      Pandas NumPy
      Feature Engineering
    MLOps & Deployment
      Docker
      CI/CD Pipelines
      MLflow
      Weights & Biases
    Programming
      Python
      SQL
      R
      C++
      Bash
    Responsible AI
      SHAP LIME
      Bias Detection
      Fairness Metrics
      Explainability

Duration: Aug 2025 – Aug 2027 | Location: Gainesville, FL

Strategic Focus: Advanced AI systems architecture, responsible AI, and enterprise-scale machine learning deployment

Relevant Coursework (In Progress):

• Deep Learning & Neural Networks
• Machine Learning for AI Systems
• Computer Vision & GANs
• Convolutional Neural Networks
• Model Optimization & Deployment

Academic Projects:

• Lab 04: GANs and Autoencoders implementation
• Lab 03: CNN architectures (2.1MB codebase, 2 commits)
• Lab 02: PyTorch fundamentals (1.3MB codebase)

Duration: Aug 2020 – May 2024 | Specialization: Internet of Things

Foundation: Strong technical foundation in CS fundamentals, IoT systems, and data structures

pie title Skill Proficiency Distribution
    "Machine Learning & AI" : 35
    "Data Engineering" : 20
    "MLOps & Automation" : 20
    "Computer Vision & NLP" : 15
    "Responsible AI" : 10

Repository: Emotion_detection_with_CNN

Technical Leadership:

• Architected real-time emotion detection using CNN and OpenCV
• Trained on FER2013 dataset (35,887 images, 7 emotion classes)
• Achieved 65-70% training accuracy with 3.5M parameter model
• Deployed with Haar Cascade face detection for real-time processing

Business Application: Foundation for patient emotional state monitoring in telehealth

Repository: A.C.I.S | Scale: 26KB codebase

Strategic Impact:

• Automated quality control for automotive assembly lines
• YOLOv8n architecture (3.2M parameters) with 99.5% mAP@0.5
• Dockerized deployment with Prometheus/Grafana monitoring
• SHAP-based explainability for quality assurance transparency

ROI: Eliminated manual inspection errors, 1000x faster than manual process

Repository: project-3-grad

Research Contributions:

• ChestMNIST: 94.92% validation accuracy, 0.8350 mean AUC-ROC (14 disease classes)
• RetinaMNIST: 57.50% validation accuracy, 0.8214 AUC-ROC (5 severity levels)
• Implemented batch normalization, dropout regularization, and class weighting
• Published 4-page IEEE-format technical report

Academic Impact: Demonstrated expertise in multi-label classification and medical AI