Available for projects

Shivani Vishwakarma.

AI Applications Developer × Agentic AI Builder

I build AI-powered applications, intelligent automation workflows, and agentic AI systems using Large Language Models, RAG architectures, and Python to solve real-world problems.

Shivani Vishwakarma
scroll

Bridging AI
& Automation

➡️

I am an IIT Madras BS Data Science student building AI-powered applications using Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), and Python. My work focuses on document intelligence, data analysis, and workflow automation.

My work sits at the intersection of AI engineering, workflow automation, and agentic systems. I build solutions using LLMs, vector databases, APIs, and orchestration frameworks to create intelligent applications that can retrieve information, reason through tasks, and take action.

I enjoy turning complex AI concepts into practical tools such as RAG-based assistants, data analytics applications, and automation workflows that solve real-world problems.

LLM Engineering Multi-Agent Systems Process Automation API Architecture n8n / Make LangChain

Technical Arsenal

...

AI Systems

Building LLM-based systems using RAG pipelines, vector databases, and reasoning workflows for intelligent information retrieval and task execution.

...

Agentic Systems

Developing AI agents capable of planning, tool usage, and multi-step reasoning to solve complex tasks autonomously.

...

Automation Engineering

Designing workflow automation systems using APIs, orchestration frameworks, and low-code tools like n8n, Make, and Zapier.

...

Python

Core language for building AI applications, data pipelines, APIs, and automation scripts using modern frameworks.

...

APIs & Integrations

Designing REST APIs, webhooks, and third-party integrations to connect services and enable scalable AI systems.

...

Data Analysis

Transforming raw data into insights using Pandas, visualization tools, and statistical analysis techniques.

...

SQL

Writing optimized queries, designing databases, and supporting data-driven applications and analytics workflows.

Also proficient in
LangChain OpenAI API Docker Git Airtable Notion API Slack API Pandas Jupyter

Selected Projects

02
AI Assistant
Visit

DocTutor AI

A document-based question answering system that enables semantic search over PDFs using embeddings and vector search.

LangChain ChromaDB Vector Search PDF QA
03
Data Science Tool
Visit

AI Data Profiler

A dataset analysis tool that automatically explores CSV data, generates insights, and summarizes patterns for faster understanding of structured data.

Python Pandas CSV Analysis Data Insights
04
Automation
GitHub

AI Lead Management System

End-to-end automation system for capturing, qualifying, storing, and following up on leads using AI-driven workflows.

n8n OpenAI Google Sheets Slack
05
Workflow Automation
GitHub

Startup Idea Evaluator

A system that evaluates startup ideas using structured AI prompts and generates consistent feedback for faster decision-making.

Make.com Google Sheets Email Automation
06
Analytics
GitHub

Sales Data Dashboard

Automated reporting system that processes sales data, extracts KPIs, and generates AI-powered insights for business performance tracking.

n8n Google Sheets OpenAI Data Analysis

The Automation Process

Every intelligent system I build follows a structured, repeatable methodology that ensures reliability, scalability, and real business impact.

01

Input & Trigger

Data enters the system from any source — form submissions, emails, webhooks, database changes, scheduled jobs, or API calls. The workflow activates intelligently based on predefined triggers and conditions.

WebhooksScheduled JobsEvent Triggers
02

AI Processing

The AI layer analyzes, classifies, and extracts meaning from incoming data. LLMs interpret context, agents consult knowledge bases, and models generate structured outputs that drive the next phase of the workflow.

LLM ReasoningRAGClassification
03

Decision Logic

Based on AI output and business rules, the system makes intelligent routing decisions. Conditions are evaluated, priority is assessed, and the workflow branches to the correct execution path — all autonomously.

Conditional LogicScoringRouting Rules
04

Action & Output

Automated actions execute across connected systems — CRMs update, emails send, Slack notifications fire, databases record, and dashboards refresh. The loop closes with logging, monitoring, and continuous improvement.

CRM UpdatesNotificationsReports

Let's Build
Something Smart

Have an automation challenge? Building an AI agent? Looking for a collaborator who can turn complex business processes into intelligent systems? Let's talk.

✓ Message sent! I'll be in touch soon.