Data Analyst | Python · Gen-AI · SQL · Tableau · Power BI | ML & NLP | B.Tech AI/ML @ VIT Chennai 2025 | Azure AI Certified

Bengaluru









B Tech, Computer Science and Engineering specialization with Artificial Intelligence and Machine Learning
Vellore Institute of Technology
12th Grade
Kathir Vidyaa Mandhir
Join Gavrish and 10k+ other professionals on
Tal is the app where bosses meet engineers & builders directly, without middlemen
Download Tal BossData Analyst | Python · Gen-AI · SQL · Tableau · Power BI | ML & NLP | B.Tech AI/ML @ VIT Chennai 2025 | Azure AI Certified
Data Analyst & AI/ML Engineer | VIT Chennai, B.Tech AI & ML (2025) | Python · SQL · Power BI · LangChain | Azure AI & Google Cloud Certified I am a recent B.Tech CS (AI & ML) graduate from VIT Chennai (CGPA: 7.92), currently working as a B2B Marketing Lead at Elytespark Innovations — where I collaborate with ML engineers and data analysts to translate AI-driven insights into business strategy. My work spans the full data stack: ETL pipeline engineering, RAG systems, AutoML platforms, model registries, and BI dashboards — all built with production-grade standards. What I bring to the table: I build systems that go beyond notebooks. From a hybrid RAG pipeline achieving 0.70+ confidence with sub-2s response time, to a Financial Index ETL ingesting 5,000+ daily records across 7 global indices with automated reconciliation reports — my projects are engineered for real-world impact. I also built an end-to-end ML Platform with a TPE-based AutoML engine, MLflow-inspired Model Registry with champion/challenger gates, and a Feature Store with PSI/KL Divergence drift monitoring — completing 50 trials in under 3 minutes. Technical Toolkit: Python · SQL · LangChain · FastAPI · FAISS · BM25 · MLflow · TensorFlow · PyTorch · Scikit-learn · Power BI · Tableau · Docker · Kubernetes · AWS · Azure AI · Google Cloud · HuggingFace · OpenCV · Pandas · Pytest Certifications: 🏅 Generative AI on Vertex AI — Google Cloud 🏅 Azure AI-900 — Microsoft 🏅 TensorFlow Developer — ZTM/Google Key Projects: 🔹 RAG-Powered Research Assistant — Hybrid FAISS + BM25 retrieval, FastAPI deployment, citation extraction 🔹 Financial Index ETL Pipeline — 7 global indices, Power BI dashboards, SLA breach logging 🔹 ML Platform (AutoML + Model Registry) — Bayesian optimisation, semantic versioning, drift monitoring I am actively seeking roles in Data Science, AI/ML Engineering, and Data Analytics — particularly in fintech, SaaS, and AI-driven product companies. 📩 gavrish1829@gmail.com 🔗 https://fantasy-decrypt1803.github.io/ 💻 github.com/Fantasy-decrypt1803
Bengaluru
aistudio
gemini
langchain
openai
Analyst at Elytespark, Mar 2026 - Present · 4 mos, Bengaluru, •Evaluated business and technical requirements to support data collection, reporting, and process optimization initiatives. •Automated repetitive data-processing tasks using Python scripts, reducing manual effort and improving workflow efficiency. •Built dashboards and visual reports using Excel, Power BI, and Tableau to track project metrics, model performance, and operational insights.
Core Member- AI/ML at Microsoft Innovations Club VITC, Jul 2022 - Apr 2025 · 2 yrs 10 mos
Ai/ML Intern at SystemTron, Oct 2023 - Nov 2023 · 2 mos
Machine Learning Intern at Feynn Labs Consultancy Services, Jun 2023 - Sep 2023 · 4 mos, • Executed data analysis in Tableau, revealing essential customer demographics that drove a 20 per cent increase in targeted marketing campaign conversions. • Segmentation led to a rise of 25 per cent in marketing efficiency and a 15 per cent rise in application rates, demonstrating the effectiveness of data-driven targeting. • Developed a resume strength analyser using NLP and deep learning, achieving 85 per cent accuracy in identifying relevant skills for job applications. • Engineered a model to evaluate skill gaps and propose targeted language upgrades, enhancing applicant fit for key positions by 20 per cent.
Machine Learning Intern at Acmegrade, Jul 2022 - Sep 2022 · 3 mos, • Designed an advanced Random Forest model for cancer prediction, achieving an increase in 30 per cent accuracy and minimising the false negatives. • Employed SMOTE (Synthetic Minority Oversampling Technique) to address class imbalance and enhanced model accuracy by 12 per cent on validation datasets. • Utilised NLP to analyse 10,000 user reviews, revealing insights that refined movie recommendations and boosted engagement by 30 per cent. • Personalised suggestions increased user interaction rates by an impressive 40 per cent, enhancing overall customer satisfaction metrics.
B Tech, Computer Science and Engineering specialization with Artificial Intelligence and Machine Learning, Vellore Institute of Technology
12th Grade, Kathir Vidyaa Mandhir