gitlost-murali Goto Github PK

followers: 17.0 following: 39.0 repos: 70.0 gists: 1.0

Name: Murali Manohar Kondragunta

Type: User

Twitter: gitlostmurali

Blog: https://gitlost-murali.github.io/blogs/

👋 Hi there, I'm Murali Manohar

My crossovers are

NLP ∩ Software Engineering
Research ∩ Industry

About me:

T-shaped engineer. Deployed a highly scalable ML application that helps ecologists in detecting bird migration patterns. Scalable upto 400 nodes.
🔭 Data Scientist with research interests in NLP. Worked on a deidentification project in Healthcare. Explored different augmentation & annotation techniques to smoothen the process. Always interested in improving under-resource setting.
🌱 I dedicate this year(Jan 2022 - 23) to be an End-to-End Data Scientist, good at Software Engineering practices, System Design, MLOps(Kubernetes, etc) & Blogging.
💬 Ask me about NLP/ML/any recent news.
📫 How to reach me: [email protected]

Murali Manohar Kondragunta's Projects

advanced-analysis-and-design-of-algorithms-course

agents-w-som

Set-of-Mark Prompting for LMMs with agents

$animation_code_for_fractional_knapsack icon$ animation_code_for_fractional_knapsack

This project is capable of generating animations for fractional knapsack problem for dynamic data.You can enter the data and click the "Start" button to see the animation

annotation-tools

Evaluate different video and image annotation tools

audioanalyzerhub

Exploratory repo. Mix of generic Video(Mostly Audio) processing techniques.

awesome-align

A word aligner based on multilingual encoders

cameratraps

Tools for training and running detectors and classifiers for wildlife images collected from motion-triggered cameras.

chai-pe-charcha

Study group, either to discuss latest happenings or to revisit the fundamental concepts.

colab_lm_pos

colab repo for pos tagging with Language model

Color detection is through opencv and numpy operations. We made this as our "Internet of Things" project.It scans an image,detects the color and moves the servo across converyor belt in a particular angle.After all this,values are uploaded to "ThingSpeak""

crowdcount-mcnn

Single Image Crowd Counting via MCNN (Unofficial Implementation)

dads-invoice-app

A web app for dad

daily-report

With approx 100 days for my next step in career, I'd like to **write** down what I read **everyday**

deconvolution-for-num-2-image

Takes a number and generates the images of those numbers. Trained on MNIST using Deconvolution Neural Networks (Fancy word for Transposed Convolution)