“Ravi has instrumental in building Albus and helped us with ideation, working on blockers and sometimes getting them fixed within 24 hours. He also served as a sounding board for new experiment and kept us updated on upcoming features in the AI world”
About
I am currently a Ai Engineer and Developer Advocate at LlamaIndex, where I integrate…
Activity
-
Last Thursday, I had the privilege of being a speaker at #SaaSCON24! As a fairly young founder (I turned 24 this week!), I normally go to events as…
Last Thursday, I had the privilege of being a speaker at #SaaSCON24! As a fairly young founder (I turned 24 this week!), I normally go to events as…
Liked by Ravi Theja Desetty
-
The latest in our series of Workflows tutorials comes from Ravi Theja Desetty, who has rebuilt our Router Query Engine using workflows. In this video…
The latest in our series of Workflows tutorials comes from Ravi Theja Desetty, who has rebuilt our Router Query Engine using workflows. In this video…
Liked by Ravi Theja Desetty
-
I wasn't sure about the buzz around the paper from Nvidia titled: LLM Pruning and Distillation in Practice. So I read it yesterday to look for new…
I wasn't sure about the buzz around the paper from Nvidia titled: LLM Pruning and Distillation in Practice. So I read it yesterday to look for new…
Liked by Ravi Theja Desetty
Experience
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Education
-
-
-
The Certificate program from International School of Engineering (INSOFE) is certified by the LTI of Carnegie Mellon University.
I was introduced to various concepts in Machine Learning,Natural Language Processing,Statistics and Probability.
Learnt concepts like Decision Trees, Random Forests, Regression methods, Linear, K-NN, Association Rules, genetic Algorithms, Monte Carlo simulations. -
-
Graduated in Electrical and Electronics engineering.
Courses :
Statistics, Data Structures and algorithms, Signals and systems, Circuit theory, Analog electronics, Digital Electronics, Computer Architecture, Microprocessors, Power Systems.
Licenses & Certifications
-
-
Sequence Models
Coursera
IssuedCredential ID https://www.coursera.org/account/accomplishments/verify/BEH8RSCLH9BE -
-
-
-
-
-
-
-
Volunteer Experience
-
Social Service
Indian Red Cross
- Present 11 years 6 months
Social Services
Courses
-
Image Analysis
-
-
Linear Algebra
-
-
Methods And Algorithms In Machine Learning
-
-
Speech Processing
-
-
Statistics And Probability
-
-
Text Mining and Natural Language Processing
-
Projects
-
Cross domain context aware Natural Language Understanding (NLU) using Attention networks
-
The goal of the project is to propose an NLU architecture which can work in a Multi-Domain scenario.
Proposed a context word based attention network to build a generalized NLU framework which will help open-domain goal-oriented task completion dialogue systems a reality. -
Answer Selection In Question Answering System - Microsoft India AI Challenge
-
This goal of the project is to select most probable answer among 10 candidate answers for a given query. Experimented with Siamese - BiLSTM network, Attentive pooling, CNN based siamese network, BERT and ELMO embeddings. Achieved an MRR score of 0.58.
-
Automatic Number Plate Recognition(ANPR)
-
Used YOLO v2 for detection of number plates on indian car vehicles.
-
Sales Forecasting
-
Forecasting daily sales and cumulative sales at division level for a leading Pharmaceutical company.
Developed R shiny application for visualisation of forecasted sales. -
Email Classification
-
Automatic routing of customer complaints of a leading bank to their respective departments.
Developed hierarchical algorithm to classify an email at three sub levels.
Used Naive Bayes algorithm for classifying an email at each level.
Honors & Awards
-
11th place (out of 121) in Rampaging Datahulk Machine Learning Hackathon
Analytics Vidhya
The task for this challenge is to predict probability whether the price for a particular stock at the tomorrow’s market close will be higher(1) or lower(0) compared to the price at today’s market close for "QuickMoney" a fastest growing hedge fund.
LB: https://datahack.analyticsvidhya.com/contest/avdatafest-rampaging-datahulk/lb -
8th place (out of 1104) in DATAFEST, month long hackathons in Machine Learning and Deep Learning
Analytics Vidhya
-
9th Place (out of 266) in DeepLearning skilltest
Analytics Vidhya
Skilltest covers Neural Networks, Convolutional Neural Networks, Recurrent Neural Networks.
LB: https://datahack.analyticsvidhya.com/contest/skilltest_deep-learning-2/lb -
8th Place ( out of 243) in Ensemble Learning Skilltest.
Analytics Vidhya
Skilltest covers different concepts in ensemble learning.
LB: https://datahack.analyticsvidhya.com/contest/skilltest-ensemble-modeling/lb -
4th place (out of 126) in MIni Data Hack
Analytics Vidhya
The challange was to predict the sales of Simple Buy for 12 months.
LB:https://datahack.analyticsvidhya.com/contest/mini-datahack-2/lb
Languages
-
Telugu
Professional working proficiency
-
Hindi
Professional working proficiency
-
English
Professional working proficiency
Recommendations received
1 person has recommended Ravi Theja
Join now to viewMore activity by Ravi Theja
-
An IT employee recently has confided something very important about upskilling which many can relate. For last 5 years, he spent on everything…
An IT employee recently has confided something very important about upskilling which many can relate. For last 5 years, he spent on everything…
Liked by Ravi Theja Desetty
Other similar profiles
Explore collaborative articles
We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.
Explore More