ChandraTeja Reddy🇮🇳
Lead AI/ML Engineer at Auditoria.ai || Creator of Ainika.AI ||
Hyderabad
6+ years of work experience
Mar 20251 yr 9 mosPresent
Hyderabad, Telangana, India
1. Successfully designed and led the development of an autonomous email processing system, ensuring seamless handling of AP invoices, accounts payable, accuracy/customer collections, and AR remittances.
2. Built an expert-based confidence scoring system, optimizing model accuracy and decision-making.
3. Implemented a feedback loop mechanism to enhance system robustness, allowing models to learn directly from user interactions and evolving data patterns.
4. Enhanced autonomous mode processing, enabling fully automated email classification, data extraction, and response generation, reducing manual intervention significantly.
May 20222 yrs 11 mosMar 2025
Hyderabad, Telangana, India
1. Authored and created Ainika AI chatbot, specializing in providing financial assistance and recognized as a master of prompt engineering.
2. Applied advanced techniques for emails intent detection, attachments, and extractions using OpenAI, Gemini, Llama2,Llama3 and Mistral.
3. Implemented data anonymization using Spacy before transmitting data/text to Large Language Models.
4. Successfully trained, fine-tuned, and deployed large language models such as Llama2,Llama3 and Mistral 7b models.
5. Possess an exceptional level of expertise with large language models, demonstrating an in-depth understanding of their capabilities.
6. Created multiple AI Assistants using OpenAI Assistants, Lang-chain, and Llama-index.
7. Chatbot session handling.
8. Transformed vision models(gemini-pro-vision,openai-vision) to detect logos, signatures, dates, and checkboxes on invoices and W9 documents.
9. Enhanced model size and reduced inference time through the implementation of lightweight models.
10. Extracted essential keywords from invoices and W9 documents, such as invoice numbers and billing addresses.
11. Implemented vision-based models as Micro-services using FastAPI.
12. Having good experience with Azure services like ADF, Data bricks, Azure functions, communications/notification services, Form Recognizer, logic-apps, and lake storages.
13. Developed automated chatbots utilizing Google Dialogue Flow CX.
14. Possess extensive knowledge of Graph Databases, specifically TypeDB.
15. Converted user natural language queries into TypeQL queries using intent and sub-intent classification.
16. Created Power Automate workflows for sending automated emails to end-users regarding the status of extracted invoice document fields/keywords.
17. Demonstrated proficiency with cloud platforms including Azure, GCP, and AWS.
18. Specialized in Graph Databases, including TypeDB and Neo4J.
Jun 20211 yrMay 2022
Chennai, Tamil Nadu, India
1. Empowered small and medium dairy farmer in India by governing financial and cattle management
associated with Computer vision driven projects.
2. For Cattle data- muzzle matching status, Cow health index, model optimizations using TF light models,
Android codes debugging, built Apk.
3. Exploited Computer Vison, Machine learning, Big data.
4. Accompanied with insurance companies like ICICI, Bajaj alliance, IFFCO tokio for Cattle trading.
5. Hands on experience in GCP
Oct 20209 mosJun 2021
Hyderabad, Telangana, India
1. Worked on multiple Azure Pipelines creations, Azure Data Lake storage Gen1 & Gen2,
Data Bricks and Azure Functions
2. Generating Data Analysis report on Microbiome data.
3. NLP problem: Extracting relationship between the entities.
4. Graql queries, Grakn knowledge graphs.
5. Solving ML Problems using Knowledge graphs.
6. Tableau, Power BI & Grankn knowledge graphs for data insights
7. Models monitoring & checking the performance
consistently using Machine learning.
8. Direct conversation with UK Clients, SME discussions & Generated Reports, closely worked with DS teams.
Jul 20204 mosOct 2020
Hyderabad, Telangana, India
Automated Disambiguation of Acronyms and Abbreviations:
Acronyms and abbreviations within electronic Technical texts are widespread and often associated with multiple senses.
Automated acronym sense disambiguation (WSD), a task of assigning the context-appropriate sense to ambiguous Technical acronyms and abbreviations, represents an active problem for medical natural language processing (NLP) systems. In this paper, fifty clinical acronyms and abbreviations with 500 samples each were studied using supervised machine-learning techniques (Support Vector Machines (SVM), Naïve Bayes (NB), and Decision Trees (DT)) to optimize the window size and orientation and determine the minimum training sample size needed for optimal performance. Our analysis of window size and orientation showed best performance using a larger left-sided and smaller right-sided window. To achieve an accuracy of over 90%, the minimum required training sample size was approximately 125 samples for SVM classifiers with inverted cross-validation. These findings support future work in Technical acronym and abbreviation WSD and require validation with other Technical texts.
Dec 20191 yr 7 mosJun 2021
ChandraTeja’s AI Stack
Education
Post Graduation Program Diploma, Data Science
International School of Engineering (INSOFE)
BTech - Bachelor of Technology, Information Technology
JB Institute Of Engineering and Technology (JBIET)
Join ChandraTeja and 10k+ other professionals on
Tal is the app where bosses meet engineers & builders directly, without middlemen
Download Tal Boss