Nandan Vallamdasu
AI Engineer
Hyderabad, India
AI Engineer blending analytical precision with creative automation to build data systems that don't just process information. They perform.
Obsessed with turning raw data into scalable intelligence through Azure, Databricks, and Python.
I craft pipelines, predictive models, and automation frameworks that learn fast, operate clean, and think ahead, engineering AI that transforms complexity into clarity.
Developer {
name: Nandan Vallamdasu;
skills: Routing boring to AI, brilliant to brains;
interests: Research, innovation, & calculated madness ;
};
interface TechStack {
devOps: Docker, Azure DevOps, GitHub Actions, n8n;
data: Azure Databricks, Data Factory, Python, SQL, Power BI;
ai: LangChain, Azure AI Foundry, Vector Databases, NLP, Sentiment Analysis;
explorationZone: Figma, MLOps, RAG, Vector Intelligence, AI Security & Ethics;
humanOps: Collaboration, Communication, Problem Solving, and the usual chaos control;
};
const craft = {
coreFocus: [
• Integrating LLMs and AI agents into real-world operations,
• Optimizing model performance and cloud efficiency,
• Designing scalable AI workflows and data architectures,
• Building intelligent automation for data-driven decisions,
• Following AI research and innovations from DeepMind, OpenAI, etc.
],
currentStack: ['Azure', 'Databricks', 'Python', 'LangChain', 'n8n']
};
const exploration = {
nowLearning: ['RAG Pipelines', 'AI Agentic Systems', 'Cognitive Orchestration'],
goal: 'Teaching AI to reason, automate, and collaborate instead of just predict.'
};
const engagement = {
availability: 'Open to collaborations in AI engineering, data automation, and research-driven systems.',
funFact: 'When not debugging data pipelines, I'm probably deep in a sci-fi binge, overanalyzing movie logic, or reading papers I only half understand.'
};
// Contact function
const contact = (type: ContactType): string => {
switch (type) {
case 'github':
return 'https://github.com/nandan2003';
case 'twitter':
return 'https://x.com/NandanV76';
case 'linkedIn':
return 'https://www.linkedin.com/in/nandan-vallamdasu';
case 'instagram':
return 'https://www.instagram.com/nandan_vallamdasu';
case 'email':
return 'nandan.vallamdasu@zohomail.in';
default:
return 'Send fax.';
}
};
HelixGPT: End-to-End CPU-Trained GPT
A fully functional 11M parameter GPT language model built and trained from scratch entirely on CPU hardware. The project spans the complete LLM lifecycle from custom tokenization and dataset sharding to training with PyTorch and cloud deployment via Docker on Azure. Note: Coldstart upto 2 minutes
Edge-CPU Inference Engine
A rigorous systems engineering suite benchmarking 8 state-of-the-art LLM architectures on commodity Azure CPUs. It quantifies trade-offs between throughput, memory pressure, and accuracy, identifying Qwen 2.5 as the Pareto-optimal solution for production-grade edge inference at 95% lower cost than GPU instances.
Multi-Model ML Web App
A centralized AI prediction platform that serves four distinct machine learning models through a unified Flask interface. Hosted on Microsoft Azure, it handles diverse tasks ranging from healthcare diagnostics to financial estimation using robust Scikit-learn pipelines. Note: Coldstart upto 5 minutes
ML & AI Portfolio: From Foundation to MLOps
A comprehensive monorepo documenting an end-to-end journey in AI. It houses over 10 projects ranging from classical regression and clustering algorithms to advanced deep learning systems (LSTMs, CNNs) and cloud-native deployments on Microsoft Azure and Docker.
GoodFoods AI Reservation Assistant
A production-oriented conversational AI agent designed for a multi-location restaurant chain. Built on a framework-free, decoupled architecture, it manages 24/7 reservations, enforces strict business logic via dynamic prompting, and ensures data sovereignty by replacing third-party booking platforms.
Formula 1 Azure Databricks Data Engineering
An end-to-end cloud data pipeline built on the Azure ecosystem. It implements a "Medallion" architecture (Bronze, Silver, Gold) using Databricks and Data Factory to ingest, transform, and analyze Formula 1 race data with automated incremental loading.
Resume: {
name: Nandan Vallamdasu,
region: Hyderabad, India,
signature: AI Engineer | Automation | Applied Research
};
Education: {
degree: Bachelor of Technology in Computer Science and Engineering,
institution: Malla Reddy Institute of Technology and Science,
duration: 2020 - 2024,
majors: Data Structures, OS, DBMS, Cloud Computing, AI & ML, C/C++
};
Certifications: [
• Google Data Analytics Professional Certificate,
• Microsoft Certified: Azure Data Fundamentals (DP-900),
• Microsoft Applied Skills: Build a NLP Solution with Azure AI Language,
• CS50x: Introduction to Computer Science (Harvard University),
• SQL (Advanced) - HackerRank
];
// # Projects
const Projects: [
{
name: 'Edge-CPU Inference Engine',
description: 'Engineered a rigorous benchmarking suite to evaluate 8 state-of-the-art LLM
architectures on commodity Azure CPUs. Quantified trade-offs between throughput, memory
pressure, and reasoning accuracy, identifying Qwen 2.5 (3B) as the Pareto-optimal solution
for edge inference at 95% lower cost than GPU instances.',
techStack: ['Python', 'C++', 'llama.cpp', 'Azure VM', 'Bash', 'Matplotlib', 'Pandas']
},
{
name: Azure Databricks Formula 1 Analytics;
description: Developed an end-to-end data engineering pipeline analyzing F1 driver and team
performance, processing incremental race data through Azure Data Factory and Databricks, and
visualizing insights via Power BI;
techStack: [Azure Databricks, Data Factory, Azure Data Lake Gen2, Python, SQL, Power BI]
},
{
name: Microsoft Fabric Bing API Sentiment Analysis;
description: Built a real-time sentiment analysis system using Bing News API and Microsoft
Fabric, automating Teams alerts for positive sentiment and visualizing news trends daily
in Power BI;
techStack: [Fabric Data Factory, Bing API, Python, Synapse Analytics, Data Activator,
Delta Lake, Power BI]
}
];
// #2 Experience
organization: Freelance Automation & Data Projects,
location: Remote,
title: Automation & Data Engineer,
duration: 2023 - Present,
achievements: [
• Developed workflow automation systems using n8n & Power Automate to streamline repetitive
business processes.
• Designed and deployed Azure-based ETL pipelines integrating Databricks, Synapse, and Power
BI for analytics visualization.
• Created intelligent dashboards and news sentiment models using Bing API and Fabric Data
Factory.
],
organization: Prudent Autolytics,
location: Bengaluru (Remote),
title: Intern - RPA Developer,
duration: Jul-2025 - Aug-2025,
achievements: [
• Developed a PowerApps-based timesheet management system from scratch, improving internal
task tracking efficiency.
• Automated Excel-based data workflows using Python scripts, reducing manual processing time.
• Built and deployed multiple mini-process automations using Microsoft Power Automate.
]
};
// # Leadership / Extracurricular
Leadership: [
• Organized 5+ large-scale technical events at Microsoft IDC with 200+ participants and 20+
volunteers.
• Facilitated collaboration across students, mentors, and industry speakers on topics like AI,
Data Science, and Cloud Computing.
• Delivered sessions on practical applications of Azure and Power Platform in solving real-world
business problems.
• Volunteered as a guest coordinator for the Isha Insight program, engaging with business leaders
including the Ex-Chairman of ISRO and CBO of Mahindra Holidays.
• Contributed to Isha Insight program success through meticulous coordination, hospitality
management, and cross-team collaboration.
];
// #4 Skills
Skills: {
technical: [
Python, SQL, Azure Databricks, Data Factory, Synapse Analytics,
Fabric Data Factory, Power BI, n8n, Power Automate, Azure AI,
Entra ID, RBAC, PaaS, Dataverse, HTML, DevOps, DSA
],
soft: [
Team Management, Leadership, Communication, Problem Solving,
Collaboration, Critical Thinking, Stakeholder Management, Attention to Detail
]
};
// #5 Interests
Interests: Data, AI, Automation, Cloud, Research, Innovation, and a bit of chaos;
Under Development