About
Activity
-
Today, Lattice made history: We became the first company to give digital workers official employee records in Lattice. This marks a huge moment in…
Today, Lattice made history: We became the first company to give digital workers official employee records in Lattice. This marks a huge moment in…
Liked by Manish Varma Datla
-
Notion AI now includes GPT-4, no outside tools or extra fees Chat privately and securely with Notion AI. Ask about historical events or new ideas…
Notion AI now includes GPT-4, no outside tools or extra fees Chat privately and securely with Notion AI. Ask about historical events or new ideas…
Liked by Manish Varma Datla
Experience & Education
Volunteer Experience
-
AI/ML Lead
CMU Tech & Entrepreneurship
- 1 year 1 month
Education
https://www.tartanx.co/machine-learning
Publications
-
Bench Marking of Classification Algorithms: Decision Trees and Random Forests – A Case Study using R
IEEE International Conference on Trends in Automation, Communication and Computing Technologies
The paper focuses on comparing the performances of two machine learning algorithms - Decision Trees and Random Forests. Through the course of the paper, a comparison is made of classification results of these two algorithms, for classifying data sets obtained from Kaggle's Bike Sharing System and Titanic problem. The paper concludes that for all classification problems Decision Trees is handy with small data sets i.e. less number of instances and Random Forests gives better results for the same…
The paper focuses on comparing the performances of two machine learning algorithms - Decision Trees and Random Forests. Through the course of the paper, a comparison is made of classification results of these two algorithms, for classifying data sets obtained from Kaggle's Bike Sharing System and Titanic problem. The paper concludes that for all classification problems Decision Trees is handy with small data sets i.e. less number of instances and Random Forests gives better results for the same number of attributes and large data sets i.e. with greater number of instances.
-
Data Mining Classification Techniques Applied to Analyze the Impact of Ambient Conditions on Aero Engine Performance – A Case Study Using Xlminer
IEEE International Conference on Electrical,Computer and Communication Technologies 2015
Efficiency and thrust output of an Aero Engine have a huge bearing and vary drastically based on the ambient conditions at airports. The amount of these variations greatly affects thrust, fuel consumption, surge margins, temperature management across the cross section of the gas turbine fluid flow path. Data mining techniques have been used in the aeronautical industry for many years and have been known to be effective. In order to solve problems such as, this study concentrates on building a…
Efficiency and thrust output of an Aero Engine have a huge bearing and vary drastically based on the ambient conditions at airports. The amount of these variations greatly affects thrust, fuel consumption, surge margins, temperature management across the cross section of the gas turbine fluid flow path. Data mining techniques have been used in the aeronautical industry for many years and have been known to be effective. In order to solve problems such as, this study concentrates on building a hybrid methodology, combining data mining techniques such as association rules and classification trees for creating patterns to aid in better performance management of aero engines. The methodology is applied to real-world parameters data collected from Forecast Systems Laboratory Radiosonde Database and the results are evaluated by comparing them with other techniques. The performance of the aero engine and in turn the airplane is accounted by analyzing various parameters like Pressure, Temperature, Wind direction and Wind Speed at different altitudes of the aircraft as it takes off and lands at the airports.
This methodology is expected to help Aeronautical Engineers to make a faster and more accurate prediction of the aero engines and aero planes performance. -
MED+BOOK:An Encyclopedia Of your Health History
International Journal of Engineering Research and Technology
In today’s world of ever increasing amounts of information, hospitals and medical groups must continually find ways to manage the myriad information that is gathered on patients. MED+YOU not only helps to manage this myriad patient information, but it also educates patients about health information technologies so that they are in a better position to monitor their health, attain desired health goals and manage their health services. MED+YOU also possesses several facets which include…
In today’s world of ever increasing amounts of information, hospitals and medical groups must continually find ways to manage the myriad information that is gathered on patients. MED+YOU not only helps to manage this myriad patient information, but it also educates patients about health information technologies so that they are in a better position to monitor their health, attain desired health goals and manage their health services. MED+YOU also possesses several facets which include self-management of health, patient doctor communication, improvement of health awareness and outcomes, patient satisfaction with their healthcare provider and assessment of accessibility and usability factors of the personal health records. Subsequently, the MED+YOU improve sustainability and efficiency of patient-centered care to be provided among healthcare networks and clinic partners throughout the globe. MED+YOU would function as a universal portal where a person would have his or her own profile with a login ID and a password. It will also contain a home page where information about health and related issues keep getting updated. We also get to form groups and follow health related pages and people with similar problems.
Patents
-
System and Method for Tracking Baggage
Filed IN 201641005683
The present invention relates to a system and method for tracking baggage and more particularly, to a push system and a method for aiding passenger to track real time status of his/her bag anytime during travel.
Projects
-
Khojo
Part of the team that worked on the problem of: “ensuring healthcare services
within 10 km for individuals living in rural India using m-healthcare”. To
solve this, we built a scalable, cost-effective technology for routing rural
patients to appropriate healthcare facilities using Asha workers.Other creators -
SAE Hyd & KGNMT
Developed the website, both the frontend and the backend. The website has been evolved by customizing a template.The backend for the site was completely restricted to PHP.
-
Cross referencing Recommendation Engine
-
The recommendation engine was built with an intent to provide more personalized recommendations to the user. The recommendation system is unique in the way that it predicts the food the user might want to have ,on the basis of what movies and music the user likes or predict the movie he might want to watch on the basis of what food and music the user likes and similarly in the case of music. The Application collects 38 parameters for user, and each of these parameters are categorized into…
The recommendation engine was built with an intent to provide more personalized recommendations to the user. The recommendation system is unique in the way that it predicts the food the user might want to have ,on the basis of what movies and music the user likes or predict the movie he might want to watch on the basis of what food and music the user likes and similarly in the case of music. The Application collects 38 parameters for user, and each of these parameters are categorized into either of the three categories I.e context-based,behavior-based and instinctive-based parameters. Furthermore,these parameters are cross-referenced with a graph built upon open data sets of music,movie and food to generate the relevant recommendations for the user.
.The was built built on top of Azure Machine Learning Studio and Open Data. The application functions as both a Business to Business(B2B) model and Business to Client(B2C) model,under the B2B model existing applications can integrate themselves with the recommendation engine via an API and the B2C model focuses on directly interacting with the user, where the user can directly come on to the Android App and identify which movie,music and food would he prefer at that point of time.Other creators -
Statistical approach for image classification
-
The project was based in the domain of Image Processing and the project was aimed at providing a custom classifier which is able to classify preprocessed textural images. I proposed a simple albeit
effective classifier which utilizes the textural features extracted from the images to compare and analyze the image features. The proposed algorithm makes use of a statistical approach where an image texture is seen as a quantitative measure of the arrangement of intensities in a region.Other creators
Honors & Awards
-
Winner for BloodChain at GS1 US Trace-a-thon Hackathon
GS1 US
The requirement for blood in the world is increasing as the population and life expectancy are going up. At the same, the trust in the existing system of blood collection, processing, usage is dwindling making lesser people come out and donate blood. The blood donors don't know what happens with their blood and how it is being used.
BloodChain, using blockchain and GS1 traceability standards is bringing trust, traceability, interoperability and seamless data sharing across the entire…The requirement for blood in the world is increasing as the population and life expectancy are going up. At the same, the trust in the existing system of blood collection, processing, usage is dwindling making lesser people come out and donate blood. The blood donors don't know what happens with their blood and how it is being used.
BloodChain, using blockchain and GS1 traceability standards is bringing trust, traceability, interoperability and seamless data sharing across the entire supply chain.
Press: https://markets.businessinsider.com/news/stocks/team-bloodchain-wins-gs1-us-trace-a-thon-for-its-innovative-traceability-solution-1028615627 -
Winner for Travel4Good at CMU Start-Up Challenge
Carnegie Mellon University
Travel4Good is a platform that empowers enterprising entrepreneurs in underdeveloped countries to provide unique travel experiences to tourists. We aim to disrupt the 2.5 billion per year travel market by providing unique off-beat travel experiences and helping bridge the socio-economic gap in some of the most economically challenged countries around the world.
-
Winner for Chain Connection at Blockchain Hackathon - Enterprise Ready
San Francisco Blockchain Club
Built the product "Chain Connection", to ease the employee onboarding process at enterprises by leveraging ethereum blockchain and smart contracts.
-
Winner for "Most Original Use Case" at BitClave Blockchain Developer Hackathon
BitClave
Built the product "Data Service (for BitClave and Base)". Data Service expand's the capability of BitClave/Base to serach on documents and general data-sets (beyond Search History and Wallet Wealth).In addition, we provide 3rd party Service that provides certificates, document management and interest extraction and uploads data to Base through API and documentations to IPFS.
-
Winner for a game "Run for your Life" - A game conceptualized around the concept of "Student of the Markets (Finance)"
Blackrock India
The game was conceptualized to provide users/individuals with a better ability to comprehend the financial markets. The game has its genesis with "Temple Run" coupled with concepts of the financial domain.
-
Winner for Fin Assistant - A chatbot in the Financial domain
Unisys Cloud 20/20 Y8
The chatbot Fin Assistant (Financial Assistant) is a one stop solution to learn and make informed decisions about companies and investments related to them.
-
Winner of Coding Milestone at Acad Accelerator
Microsoft
Won the Coding milestone for our recommendation engine framework - "Multi-Dimensional Cross Category Recommendation Engine Framework for Human Preferences", as part of the three-month mentoring program by Microsoft.
-
Finalist and Honorable Mention for Android based IoT System
CISCO iDeate - The Great Digital India Challenge
The Application is an Intelligent Baggage Handling system using RFID. With an integrated Analytics platform the application offers perfect end to end sorting,tracking and tracing technology.The end product fis categorized into two end forms the B2B and the B2C applications.Our B2C mobile application interacts with the passenger and enables the passenger to track the baggage status at all points during journey.The B2B Analytics platform provides in depth analysis for baggage associated with each…
The Application is an Intelligent Baggage Handling system using RFID. With an integrated Analytics platform the application offers perfect end to end sorting,tracking and tracing technology.The end product fis categorized into two end forms the B2B and the B2C applications.Our B2C mobile application interacts with the passenger and enables the passenger to track the baggage status at all points during journey.The B2B Analytics platform provides in depth analysis for baggage associated with each passenger, thereby trying to preempt the likelihood of a baggage mishandling scenario. The baggage handling system is an end product for airport authorities/airline services and a value addition service for the passengers.
-
Finalist for Multi-Dimensional Cross Category Recommendation Engine Framework for Human Preferences
MICROSOFT Imagine Cup India National Finals 2016
-
Finalist for IOT based Smart Recommendation Engine - 8Ball
UNISYS Cloud 20/20 V7.0
8Ball is an IoT based smart recommender system. Built with an intent to provide personalized recommendations to the user, our Multi-Dimensional Cross Category Recommendation Engine Framework for Human Preferences lies at the core of our ability to provide these recommendations. Unlike most recommender systems we are nonlinear i.e. we are unique in the way we predict the food the user might want to have on the basis of what movies and music the user likes or predict the movie he might want to…
8Ball is an IoT based smart recommender system. Built with an intent to provide personalized recommendations to the user, our Multi-Dimensional Cross Category Recommendation Engine Framework for Human Preferences lies at the core of our ability to provide these recommendations. Unlike most recommender systems we are nonlinear i.e. we are unique in the way we predict the food the user might want to have on the basis of what movies and music the user likes or predict the movie he might want to watch on the basis of what food and music the user likes and similarly in the case of music. To establish a context associated with the user we built a ‘Mind Map’ which consists about 45 parameters associated with the user, ranging from the user heartbeat to location, speed etc. each of which are categorized into either of the three categories i.e. contextual-based, behavioral-based and instinctive-based data parameters. Furthermore these parameters are cross-referenced with an open data graph built upon open data sets of music, movie and food to generate the relevant recommendations for the user, with an integrated feedback system we at all instincts of time establish a more personalized context in regards to the user.
Built on top of Azure ML, we categorize our end products into the B2B and the B2C domains. Under the B2B model third party platforms/products can avail our services via our 8Ball API. In the B2C context we interact directly with end user via our mobile phone application.
-
Winner for an Android based recommendation engine - RecommendationE.
MICROSOFT Code.fun.do
The recommendation engine was built with an intent to provide more personalized recommendations to the user. The recommendation system is unique in the way that it predicts the food the user might want to have ,on the basis of what movies and music the user likes or predict the movie he might want to watch on the basis of what food and music the user likes and similarly in the case of music. The Application collects 38 parameters for user,and each of these parameters are categorized into either…
The recommendation engine was built with an intent to provide more personalized recommendations to the user. The recommendation system is unique in the way that it predicts the food the user might want to have ,on the basis of what movies and music the user likes or predict the movie he might want to watch on the basis of what food and music the user likes and similarly in the case of music. The Application collects 38 parameters for user,and each of these parameters are categorized into either of the three categories I.e context-based,behavior-based and instinctive-based parameters. Furthermore,these parameters are cross-referenced with a graph built upon open data sets of music,movie and food to generate the relevant recommendations for the user.
.The was built built on top of Azure Machine Learning Studio and Open Data. The application functions as both a Business to Business(B2B) model and Business to Client(B2C) model,under the B2B model existing applications can integrate themselves with the recommendation engine via an API and the B2C model focuses on directly interacting with the user, where the user can directly come on to the Android App and identify which movie,music and food would he prefer at that point of time. -
Finalist for IOT based Android application for Smart Medication - MedING
INK Makers
The application is developed targeting the Drug and the Pharmaceutical setup in India. With the ever increasing use of Drugs, the patients today end up spending more on the prescribed drugs rather than cheaper alternative generics. The Application is aimed at suggesting cheaper alternative equivalent generics for a drug and at the same time suggest drugs/generics for a particular condition. The application is completely built on open data, where the data obtained is interpreted and visualized…
The application is developed targeting the Drug and the Pharmaceutical setup in India. With the ever increasing use of Drugs, the patients today end up spending more on the prescribed drugs rather than cheaper alternative generics. The Application is aimed at suggesting cheaper alternative equivalent generics for a drug and at the same time suggest drugs/generics for a particular condition. The application is completely built on open data, where the data obtained is interpreted and visualized on the analytics platform Socrata. My case study, on the usage of generics instead of prescribed drugs has shown a decrease in cost of medication for an individual by at least 20% per month.
-
Winner for IOT based Android application
HONEYWELL TekUnite
The Application is built as a part of an IOT product for Smart Airports.
The Application aims at aiding the passenger track the status of his/her bag anytime during the journey using the Application. Using the RFID baggage tracking technology, the application aims to provide virtually perfect end-to-end sorting, tracking and tracing of baggage. RFID tags are pasted on the baggage and then are linked to the boarding card in the form of a QR Code or BarCode. The information from the RFID is…The Application is built as a part of an IOT product for Smart Airports.
The Application aims at aiding the passenger track the status of his/her bag anytime during the journey using the Application. Using the RFID baggage tracking technology, the application aims to provide virtually perfect end-to-end sorting, tracking and tracing of baggage. RFID tags are pasted on the baggage and then are linked to the boarding card in the form of a QR Code or BarCode. The information from the RFID is stored on the server and simultaneously displayed on the Application.
Languages
-
English
Full professional proficiency
-
Hindi
Professional working proficiency
-
French
Limited working proficiency
-
Telugu
Native or bilingual proficiency
More activity by Manish Varma
-
📈 Exciting to see the progress from Dharmesh Shah, the visionary builder behind HubSpot, with his groundbreaking project, Agent.ai! 🚀 🗺 At…
📈 Exciting to see the progress from Dharmesh Shah, the visionary builder behind HubSpot, with his groundbreaking project, Agent.ai! 🚀 🗺 At…
Liked by Manish Varma Datla
-
Throwback Sunday to Hugging Face’s first ever parisian office (2016-17) Just above Higuma 🍜 Then we stepped up to STATION F when it opened 🔥
Throwback Sunday to Hugging Face’s first ever parisian office (2016-17) Just above Higuma 🍜 Then we stepped up to STATION F when it opened 🔥
Liked by Manish Varma Datla
-
ChatGPT is the fastest-growing consumer app in the history of the internet while "OpenAI Mafia" is the fastest-growing group of founders &…
ChatGPT is the fastest-growing consumer app in the history of the internet while "OpenAI Mafia" is the fastest-growing group of founders &…
Liked by Manish Varma Datla
-
Congratulations to Rishit Khare and the team on successfully launching your first software product. It is remarkable to see such a beautiful…
Congratulations to Rishit Khare and the team on successfully launching your first software product. It is remarkable to see such a beautiful…
Liked by Manish Varma Datla
-
After Azure exit last month, we’ve now fully exited google maps. We used to spend ₹100 cr a year but we’ve made that 0 this month by moving…
After Azure exit last month, we’ve now fully exited google maps. We used to spend ₹100 cr a year but we’ve made that 0 this month by moving…
Liked by Manish Varma Datla
-
Sparse Lexical and Expansion models, or SPLADE, employ a very cool algorithm to enhance the performance of sparse information retrieval algorithms…
Sparse Lexical and Expansion models, or SPLADE, employ a very cool algorithm to enhance the performance of sparse information retrieval algorithms…
Liked by Manish Varma Datla
-
Venture has been a more fulfilling career than I could've ever imagined. Grateful for my teammates at Bain Capital Ventures, the founders I'm lucky…
Venture has been a more fulfilling career than I could've ever imagined. Grateful for my teammates at Bain Capital Ventures, the founders I'm lucky…
Liked by Manish Varma Datla
-
Our team is growing. Welcome Yash Agarwal!
Our team is growing. Welcome Yash Agarwal!
Liked by Manish Varma Datla
-
Thank you Redpoint & Nasdaq for the recognition. We'll continue to work hard and climb our way to the top.
Thank you Redpoint & Nasdaq for the recognition. We'll continue to work hard and climb our way to the top.
Liked by Manish Varma Datla
-
THE AI INNOVATOR’S DILEMMA A huge % of SaaS companies that were founded ~5-8 years ago and got to $10-50M ARR, are now building and upselling a new…
THE AI INNOVATOR’S DILEMMA A huge % of SaaS companies that were founded ~5-8 years ago and got to $10-50M ARR, are now building and upselling a new…
Liked by Manish Varma Datla
-
Critical find by Qualys. There is already a POC on the GitHub and detection is going to be not fun. 😬 Affected OpenSSH versions: OpenSSH versions…
Critical find by Qualys. There is already a POC on the GitHub and detection is going to be not fun. 😬 Affected OpenSSH versions: OpenSSH versions…
Liked by Manish Varma Datla
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