Pooja Chandrashekara

AI Researcher

Arlington, Virginia, United States2 yrs 9 mos experience

Most Likely To SwitchAI ML Practitioner

Key Highlights

Designed a RAG system boosting user engagement by 45%
Secured investment driving 30% revenue increase
Developed a QA system improving retrieval accuracy by 25%

Stackforce AI infers this person is a SaaS expert specializing in AI-driven solutions for user engagement and data processing.

Contact

poojachandrashekar0608@gmail.com LinkedIn

Skills

Core Skills

Artificial Intelligence (ai)Large Language Models (llm)Retrieval-augmented Generation (rag)Natural Language Processing (nlp)Data Architecture

Other Skills

Amazon Web Services (AWS)Analytical SkillsBERTBERT (LLM)BERT (Language Model)Big DataC (Programming Language)C++ColabData AnalysisDatabase DesignEngineeringEnvironmental Impact AssessmentGPT-3Generative AI

About

With a Master's degree in Data Science from George Washington University, I am currently applying my expertise as an AI Engineer at Travana. My core competencies lie in RAG and large language models, which have significantly improved our platform's user engagement and reduced customer churn. At Travana, our team has designed a Retrieval Augmented Generation (RAG) system that not only surpassed traditional chatbots but also secured crucial investment by demonstrating the platform's capabilities. My approach prioritizes collaboration, employing advanced technologies to enhance user interaction, which aligns with our organization's culture of innovation and commitment to driving growth through cutting-edge solutions.

Experience

2 yrs 9 mos

Total Experience

1 yr 4 mos

Average Tenure

1 yr 8 mos

Current Experience

The world bank

AI Engineer

Oct 2024 – Present · 1 yr 8 mos · Washington, District of Columbia, United States · On-site

Artificial Intelligence (AI)Large Language Models (LLM)Generative AIGenerative AI ToolsRetrieval-Augmented Generation (RAG)

Intripid (formerly travana)

AI Engineer

Sep 2023 – Oct 2024 · 1 yr 1 mo · Texas, United States · Hybrid

Designed and implemented a Retrieval Augmented Generation-powered platform for the travel industry using Langchain and ChromaDB, leading to a 45% increase in user engagement and 2000 active users per month.
Reduced customer churn by 21% by designing and deploying an advanced RAG system, effectively surpassing traditional chatbots and competitors by enhancing user interaction capabilities.
Leveraged advanced RAG technology to secure investment, driving a 30% increase in company revenue by demonstrating enhanced platform capabilities and future potential.
Conducted extensive research and analysis on various RAG technologies, evaluating their uses and limitations to optimize platform performance, resulting in a 20% improvement in system efficiency.
Engineered a hybrid RAG system integrating knowledge graphs and vector databases, enhancing data processing and retrieval efficiency, achieving a 25% increase in data retrieval speed.
Worked with cross-functional teams to integrate the RAG system into existing travel services, improving user experience and satisfaction, evidenced by a 15% increase in positive user feedback.

Artificial Intelligence (AI)RAGLeadershipKnowledge Graph EmbeddingsMachine LearningLarge Language Models (LLM)+3

Syntegral

Data Science and NLP Intern

Jan 2023 – Apr 2023 · 3 mos · New York, United States

Increased information retrieval accuracy by 25% by developing a Question Answering (QA) system using GPT-3 with embeddings and retrieval techniques, focusing on Wikipedia articles about Impact Investing and ESG.
Reduced response errors by 30% through the implementation of strategies to prevent hallucination and improve GPT-3's accuracy, including prompt engineering and the addition of explicit contextual information.
Enhanced data processing efficiency by 20% by preprocessing a document library into semantically related sections and creating embedding vectors for each section with the OpenAI Embedding API.
Boosted question-answering precision by 35% using the OpenAI Completions API to generate responses tailored to user queries with the most relevant contextual information.
Magnified GenAI framework's reliability and trustworthiness by 20% through comprehensive data experiments, developing data pipelines compliant with over 10+ international standards and taxonomies.
Spearheaded a cross-functional team to achieve a 15% increase in prediction accuracy by enhancing Environmental, Sociological, and Governance (ESG) impact analysis with statistical modeling and machine learning.

Artificial Intelligence (AI)Environmental Impact AssessmentNatural Language Processing (NLP)Problem SolvingRAGAnalytical Skills+4