Nirdosh Gandhi
Data Engineer · Azure Databricks · PySpark · ML Pipelines
📍 Bengaluru, India · 🏅 Databricks Certified Data Engineer
What I Bring
Data Engineering
Scalable, production-grade ETL pipelines on Azure Databricks using Medallion Architecture (Bronze → Silver → Gold), PySpark, and SQL — serving multi-market pharma analytics.
ML & AI Integration
M.Tech in Artificial Intelligence from DTU with hands-on experience delivering governed, audit-ready datasets that power ML model deployment in live production systems.
DevOps & Automation
End-to-end CI/CD pipelines using GitHub Actions and Databricks Asset Bundles — automating artifact packaging, deployment, and environment configuration across dev, staging, and prod.
Experience & Education
Experience
Associate - Data Engineer
Axtria - Ingenious Insights - [Bengaluru, Karnataka]
May, 2025 - Present • Website: axtria.com/
- Building a reps-recommendation platform that delivers ML-driven insights via Veeva CRM, empowering pharma sales reps to prioritise HCP engagements and improve appointment conversion rates.
- Implemented GitHub Actions CI/CD to fully automate packaging and deployment of artifacts, wheels, and notebooks from GitHub to Databricks workspaces across environments — eliminating manual Azure DevOps dependencies.
- Deployed complete workflows, job orchestration, and environment-specific configurations as code using Databricks Asset Bundles (DAB) — enabling one-command deployments across dev, staging, and prod.
- Architected a scalable, cost-optimised data platform on Azure Databricks with Unity Catalog, enforcing fine-grained data governance and access control across multiple pharmaceutical markets.
- Engineering SQL transformations and orchestrating multi-market data pipelines across Medallion Architecture (Bronze → Silver → Gold) using YAML workflows and Databricks Jobs — delivering governed, production-ready data at scale.
- Partnering with Data Science and Product teams to deliver audit-ready datasets that directly power ML model deployment in production environments.
Analyst
July, 2024 - May, 2025
- Owned end-to-end ETL pipelines using Azure DevOps, Databricks, Azure Data Factory, and PySpark — delivering reliable, production-grade data for pharma analytics across multiple business regions.
- Onboarded new markets for pharmaceutical survey data and integrated new KPIs end-to-end. Co-delivered the client organisation's first Unity Catalog (UC)-enabled product, establishing a new standard for governed data delivery.
- Built AutoQC pipelines in Azure Data Factory to enforce data quality at scale. Established CI/CD pipelines and reverse-synchronised ARM templates to align Dev with Prod — eliminating environment drift and enabling continuous safe development.
Campus Intern
Jan, 2024 - July, 2024 • (Link to Certificate)
- Analysed pharmaceutical survey data to surface actionable business insights using Python, SQL, and Microsoft Excel. Gained hands-on exposure to Azure Databricks, DBeaver, and production ETL tooling — directly contributing to live client deliverables from day one.
Project Intern
Bhaskaracharya National Institute for Space Applications and Geo-informatics (BISAG-N) - [Gandhinagar, Gujarat]
Jan, 2022 - April, 2022 • (Link to Certificate) • Website: https://bisag-n.gov.in/
- Conducted full-cycle EDA on the Indian road accident dataset using NumPy, Pandas, Matplotlib, Seaborn, and Tableau — identifying key severity patterns and visualising findings for stakeholder review.
- Trained and evaluated ML classifiers to predict accident severity (slight vs. serious): achieved 83.3% accuracy with Logistic Regression and 74.6% with Decision Tree. Deployed the application on Heroku.
- Project: fsep2022.herokuapp.com/
Summer Intern
June, 2021 - July, 2021 • (Link to Certificate)
- Built a News Summarizer web app using Django and NLP (TF-IDF word frequency algorithm) that fetches, summarises, and categorises news articles from live APIs — deployed on Heroku.
- Extended the platform to support news fetching in 12 languages based on user country preference, broadening accessibility for a global audience.
- Project: news-portal-summarizer.herokuapp.com/
Education
M.Tech, Artificial Intelligence
(Department of Computer Science & Engineering)
Delhi Technological University - [New Delhi, Delhi]
Aug, 2022 - June, 2024 • Website: dtu.ac.in/
- CGPA: 8.95/10.00
- Relevant Coursework: Machine Learning, Artificial Neural Networks, Natural Language Processing, Pattern Recognition, Advanced Computer Vision, Predictive Modelling, Blockchain, Heuristic Methods, Advanced Data Structures and Algorithms, Principles of Artificial Intelligence, Information and Network security, Advanced Database Management System
B.Tech, Computer Engineering
Birla Vishvakarma Mahavidyalaya - [Anand, Gujarat]
July, 2018 - May, 2022 • Website: bvmengineering.ac.in/
- CPI: 8.07/10.00
- Relevant Coursework: Data Structure, Design and Analysis of Algorithms, Operating Systems, DBMS, Computer Networks, C Programming, Discrete Mathematics, Theory of Computation, Compiler Design, Software Engineering, Information Security, Data Analytics and Visualization, Machine Learning, Android Programming, Basics of Embedded and IoT Systems, .NET Technology
Gujarat Secondary and Higher Secondary Education Board
• Website: gseb.org/
- HSC-12th Marks: 96.50%, Saraswati Vidyalaya - Surat, Gujarat
- SSC-10th Marks: 86.15%, Gajera Vidyabhavan - Surat, Gujarat
Tech I'm familiar with
Azure
Databricks
Python
Apache Spark
MySQL
Git
GitHub
Django
Flask
PostgreSQL
Scikit
Numpy
Pandas
MS Excel
C++
C
Java
JavaScript
Firebase
Heroku
PythonAnywhere
R
Bootstrap
PHP
Jquery
Arduino
IFTTT
OpenCv
Jira
Confluence
Projects
Road Accident Analysis & Parking Offences
End-to-end ML project on Indian road accident & NYC parking violations datasets. Built with Python, Flask, Tableau and deployed on Heroku. Achieved 83.3% accuracy with Logistic Regression and 74.6% with Decision Tree to classify accident severity.
- • Tools & Technologies:
Python, Flask, Tableau, Heroku
Skin Cancer Detection using CNN
Skin cancer detector using a fine-tuned ResNet-50 deep learning model (TensorFlow & Keras) to classify malignant vs. benign cells in dermoscopy images. Achieved 92.4% accuracy — demonstrating the potential of CNNs in medical imaging diagnostics.
- • Tools & Technologies:
TensorFlow, Keras, Python
News Summarizer Portal
Built the News Summarizer web application using Django, MySQL & NLP. Deployed it on the Heroku cloud platform. The app fetches articles of different categories from news API and display the summaries using TF-IDF algorithm. News are fetched in 12 different languages according to users’ prefilled country preference.
- • Tools & Technologies:
NLP, Django, MySQL, Heroku
Gandhi Art Drawing
Full-stack Django web app built to promote a family art business — deployed on PythonAnywhere with role-based access (admin & customer). Drives ~400 visits/month and contributed to a 40% increase in sales. Real-world impact from a personal project.
- • Tools & Technologies:
Django, Bootstrap, PythonAnywhere
Invisible Cloak
Harry Potter-style invisibility cloak using OpenCV — masks objects in real-time by HSV colour range with bitwise AND between a static background and live webcam feed. The LinkedIn post went viral with ~1,500 reactions, showcasing the fun side of computer vision.
- • Tools & Technologies:
Python, OpenCV
LRU LFU Cache Visualizer
Demonstrated cache algorithms with efficient data structures like hash maps and linked list to optimize memory management. Visualized graphics component using the C++ SFML(Simple and Fast Multimedia Library) - graphics library to illustrate the cache operations like get and put.
- • Tools & Technologies:
C++, Data tructures and Algorithms, OS, SFML
Mini Projects
Gesture Controller
Executable setup to control mouse & keyboard events with human gestures using Computer Vision. Used haarcasacde classifier for face detection, cvzone’s hand tracking Module, imutils to fetch eye landmarks from dlib shape predictor.
-
• Tools & Technologies:
Tkinter, Haarcascade, CVZone, Dlib
AI Email Classifier
Classified unread mails from the authorized gmail account using IMAP4 protocol. Used GridSearchCV, TfdfTransformer for feature extraction & SGDClassifier to train the model. Mails are classified into 20 different categories. Achieved 76.6% accuracy.
-
• Tools & Technologies:
SGDClassifier, Flask, IMAP4 protocol
G-Home Assistant
NodeMCU based model which control our home appliances on used with IFTTT or AdaFruit service. On giving the voice command to Google Assistant or triggering the button on Blynk application, it toggles the home appliances.
- • Tools & Technologies:
IFTTT, Blynk, NodeMCU
Js Task Manager
A To-do task manager which stores user data in localStorage of browser using JavaScript, jQuery and JSON. It has some basic CSS animation & also have searching-sorting bars & task manager tabs for doing specific operations.
- • Tools & Technologies:
JavaScript, jQuery, Bootstrap
Car Parking System
Online Simulation of Car Parking System. It display the number of Vacant parking positions. It works with Micro servo motor, Potentiometer, Push buttons and LED 16x2. You can simulate it online using given demo link.
- • Tools & Technologies:
Tinkercad, IOT
Training & Professional Certifications
Achievements
✽ Axtria Bravo Award (Mar 2025): For delivering quality work under tight deadlines and earning client appreciation.
✽ Qualified GATE CSE 2022 examination conducted by IIT Kharagpur.
✽ Selected in the regional round of Smart Gujarat for New India Hackathon (SGIH 2019-20).
✽ Runner-up in a college-level competitive programming contest.
✽ Recognized by Google for creating Actions on Google, rewarded with a Google Home and goodies.
List of Publications • (Google Scholar Profile) 
Beyond pixels: The synergy of vision and language in image captioning
• Link to Publication
• Link to Publication
Industrial IoT: Overview, Enabled Technologies, Recent Case Studies & Challenges
• Link to Publication
• Link to Publication
Coding Profiles
GitHub Activity
Live contribution activity bull; github.com/ngandhi369
Extracurricular & Co-curricular Activities
Tenure: Sep, 2022 - Dec, 2024
• Assisted over 70 undergraduate students in understanding and implementing programs related to Operating Systems and the C language.
• Assisted over 70 undergraduate students in understanding and implementing programs related to Operating Systems and the C language.
Tenure: Jul 2019 - Mar 2020
• A technical community dedicated to assisting colleagues and junior members in exploring new tools and technologies in computer science.
• Organized various workshops, expert talks and programming contests for them.
• A technical community dedicated to assisting colleagues and junior members in exploring new tools and technologies in computer science.
• Organized various workshops, expert talks and programming contests for them.
Tenure: May 2019 - Present
• Contribute information, reviews and photos about local business and edit missing information on Google Maps, helping users make informed decisions since May 2019.
• Achieved Local Guides Level 6.
• Contribute information, reviews and photos about local business and edit missing information on Google Maps, helping users make informed decisions since May 2019.
• Achieved Local Guides Level 6.
• Successfully Cleared Intermediate Drawing Grade Examination in 2015.
• Link: drawing_intermediate/
• Learning Piano by myself on casio and sometimes on phone too.
• Link: drawing_intermediate/
• Learning Piano by myself on casio and sometimes on phone too.
Tenure: July 2018 - July, 2020
• Attended many seminars & workshops on Machine learning, System Design, AWS.
• Participated in different competition and also attended expert & alumni talks.
• Webiste: bvmengineering.ac.in/CSI/
• Attended many seminars & workshops on Machine learning, System Design, AWS.
• Participated in different competition and also attended expert & alumni talks.
• Webiste: bvmengineering.ac.in/CSI/
Contact
Location:
Bengaluru, India
Email:
gandhinirdosh@gmail.com
Lotus feet of H.D.H. Prabodhjivan swamiji