Jahnavi Dingari

AI PRODUCT ENGINEER| Generative AI | Agentic AI | Deep Learning | LLMs

Hyderabad, India

jahnavidingari04@gmail.com

About

Jahnavi Dingari is an AI Engineer specializing in Generative AI, Agentic AI, and production-grade AI systems. A Computer Science and AI graduate from Amrita Vishwa Vidyapeetham, she combines expertise in machine learning, deep learning, Generative AI , Agentic AI and scalable backend engineering to build intelligent, real-world solutions. Beyond technical development, Jahnavi brings demonstrated leadership experience through her experience in college clubs, event management, and sports, alongside published AI research in international conferences, she brings a strong blend of technical expertise, innovation, and team leadership.

Skills

Python

Java

C++

SQL

MongoDB

Data Structures and Algorithms

Object-Oriented Programming

Machine Learning

Deep Learning

Reinforcement Learning

Natural Language Processing (NLP)

Generative AI

Large Language Models (LLMs)

AI Agents

Agentic AI

Speech Processing

Data Science

Cybersecurity

Full-Stack Development

Cloud Computing

AWS

Microsoft Azure

Google Cloud Platform (GCP)

DevOps

Git

Software Development Life Cycle (SDLC)

Linux

LangChain

Langgraph

Crew Ai

Multi-Agent Systems

AI Agent Development

LLM Orchestration

Compliance

ISO

Hippa

Soc2

MCP

FastAPI

Experience

AI PRODUCT ENGINEER

TRANSNATIONAL AI

Feb 2025 to Present

Architected a multi-agent company-intelligence platform using Python, FastAPI, asyncpg, and PostgreSQL to discover, validate, and track 123K+ companies across the US and India, covering 9 federal certification categories (MBE/WBE/SBE/DBE/HUB/VBE/SDVBE/LGBTBE/DOBE) and ISO, HIPAA, SOC 2, PCI-DSS, and SOX compliance data.
Independently developed specialized AI agents for HIPAA, ISO, and regulatory compliance, alongside fraud-detection and content-moderation agents, leveraging agent orchestration, tool calling, event-driven workflows, asynchronous execution, risk scoring, explainable outputs, and human-in-the-loop review for scalable, auditable decision-making.
Optimized backend analytics through pre-aggregated summary tables, state-level drill-down filtering, and reusable cross-source data normalization, eliminating runtime GROUP BY queries on page load; delivered live dashboards for real-time KPIs, fraud trends, compliance violations, and risk-ranked review queues, while conducting security and performance reviews across ~3K LOC of asynchronous Python code.

Projects

MarketCrew – Agentic AI Marketing Workflow Orchestration

Designed a multi-agent AI system that automates marketing workflows through collaborative agents for content planning, generation, and validation. Implemented role-based agent orchestration to improve content consistency, task reliability, and execution efficiency.

Python

Agentic AI

CrewAI

LLMs

Generative AI

Equity Research Tool – Rocky Bot

Developed a Retrieval-Augmented Generation (RAG) application using LangChain for semantic question answering over financial documents. Implemented text chunking, embeddings, FAISS vector search, and LLM-based response generation to provide context-aware answers.

Python

LangChain

RAG

LLMs

FAISS

NLP

Generative AI

Flood Detection on Satellite Imagery

Developed a deep learning-based flood detection system using satellite imagery and DeepLabV3+, U-Net, and Attention U-Net architectures. Achieved 96% accuracy with DeepLabV3+ through advanced preprocessing and data augmentation, supporting scalable flood monitoring and disaster response.

Python

Deep Learning

DeepLabV3+

U-Net

Attention U-Net

Image Processing

Human Scream Detection for Crime Prevention

Developed an audio classification system to distinguish screams from non-scream sounds for potential crime monitoring applications. Evaluated SVM, Random Forest, and CNN models using MFCC audio features, achieving 95% detection accuracy to support real-time alerts and faster emergency response.

Python

Machine Learning

Signal Processing

SVM

Random Forest

CNN

MFCC

Stock Trading Agent Using Deep Reinforcement Learning

Developed a deep reinforcement learning-based stock trading agent that learns trading strategies by interacting with a simulated market environment. The system applies reinforcement learning techniques to analyze market data and make trading decisions, with the goal of improving portfolio performance through adaptive decision-making.

Python

Deep Reinforcement Learning

Machine Learning

Stock Market Analysis

Data Processing

Financial Dashboard Using MERN Stack and Machine Learning

Developed a full-stack financial dashboard integrating the MERN stack with machine learning to visualize financial data and support data-driven analysis. Built an interactive web interface with React, backend services using Node.js and Express.js, and MongoDB for data management. Integrated machine learning capabilities for financial data analysis and predictive insights.

MongoDB

Express.js

React.js

Node.js

Machine Learning

JavaScript

Education

Bachelor of Technology (B.Tech) in Computer Science and Artificial Intelligence

Amrita Vishwa Vidyapeetham · A+

Oct 2021 to Aug 2025

Certifications

Software Engineering Virtual Experience Program.