May 2023 – Present · 2 yrs 10 mos · Remote

• Driving developer adoption and engagement by producing technical content, presentations, blog posts, tutorials, sample code, and social media content to increase awareness, and educate developers in machine learning and AI.
• Establishing credibility with developers and developer communities as a thought leader in machine learning.
• Designing and representing the company in hackathons, establishing the objectives and providing guidance throughout.
• Creating technical and marketing assets, such as demos and videos, while managing the community to help developers utilize and adopt Clarifai.
• Advocating internally for the community’s needs, collaborating with cross-functional teams to influence features and technology.
• Editing videos, producing content, and providing voiceovers for tutorials and presentations created by other teams.
• Working with the Docs team, sharing user feedback and consistently updating the documentation.
• Running a weekly developer newsletter that shares tutorials, new AI models, and tips with developers.
• Collaborating with the Product team on monthly releases, participating in early testing, sharing feedback, and working on the release email and release blog.
• Partnering with the Sales team to build effective demos and UI modules needed for customer presentations.

Nov 2022 – Apr 2023 · 5 mos · United States · Remote

• - Steamship is building Heroku for NLP Services: the fastest way to create and deploy the pieces of your product that depend upon natural language understanding.

Jul 2022 – Nov 2022 · 4 mos · Tel Aviv District, Israel · Remote

May 2022 – Jul 2022 · 2 mos · Remote

• I've been selected as one of the 50 collaborators to work on the Project "Identifying Power Infrastructure Through GIS and Machine Learning to Accelerate the Green Transition".
• Project aims to tackle the problem of identifying power grid lines by applying GIS and machine learning
• Lead the Model-Building team which can best detect the pylons, substations and power stations using the SoA (state of the art) machine learning models like Yolo V5 and Yolt V5.
• Followed an Iterative Approach and worked closely with the data collection and preprocessing team to achieve the best results and to satisfy the client's requirements of detecting the pylons on different zoom levels, crowded residential area, forest area, water and also to not detect Windmills as Pylons.

Apr 2022 – May 2022 · 1 mo · Hyderabad

• Even though hospital staff spends a considerable amount of time consistently answering the same questions. Many Patient's queries go unattended
• Project: Text-based healthcare chatbot supporting admitted patients in Hospitals
• Tasks : Data Collection | Exploratory Data Analysis & Visualization | Data Preprocessing | Model Building | Deployment
• Lead the Data Preprocessing team and worked closely with the modelling team to preprocess the data which includes filtering out relevant questions and answers for the patients, dealing with missing data, truth check of the data collected etc

Feb 2022 – Jan 2023 · 11 mos

• Global AI Hub is the Swiss-based leading global community for AI education and AI career opportunities.
• We aim to educate 10M AI talents globally.

Sep 2021 – Dec 2021 · 3 mos

• During my internship as a Machine Learning Engineer worked closely with the Cofounder of Neolen and accomplished the following tasks:
• Project 1:
• Client requirement is a Machine Learning Model that removes the original background and adds a Virtual Background to a Video Conference
• By Considering the Client deadline of 1 month, we have worked with SOA(state of the art models) like Resnet and Mobilenet V2 and delivered the Project
• Project 2:
• Worked on a Sports Application and Implemented Machine learning Models to group Cricket Players into the same category using Clustering Algorithms like K Means

Aug 2021 – Sep 2021 · 1 mo

• Worked on Market Basket Analysis Project, which predicts the customer satisfaction with a
• product which can be deduced after predicting the product rating a user would rate after
• he makes a purchase.
• Collected the data from Kaggle which includes multiple CSV files, and done data preprocessing which includes merging all the CSV files into a single one, handling missing values and misspelled words.
• Done Data Visualization which helped us to detect the important features and also correlated features and Data Selection in which we only selected the features which has high impact on the Target Variable.
• Derived new features in Feature Engineering and build multiple models like Decision Tree, and Random Forest where we got accuracy of 78% and implemented Hyper Parameter Tuning where we could improve the accuracy to 82%
• Created a Frontend using Flask and deployed the Application in Heroku