Bengaluru, Karnataka, India
16K followers
500+ connections
About
Activity
16K followers
-
Amit Manchanda posted thisHiring for PMs, SPMs, GPMs, Designers and Senior Designers for ALLEN. Please drop your CV on product.hiring@allen.in #hiring #ProductManagement
-
Amit Manchanda posted thisSomeone asked me what framework to use to evaluate the adoption of AI in different sectors. From a 30k ft, I would want to use: 1. Price Sensitivity: Sectors with tighter margins, like e-commerce and smartphones, will be rapidly embracing AI to boost efficiency and profitability. 2. User Experience: Sectors that demand good end-user experience will start to see experiments in the AI Space. 3. High Barrier to entry: Industries with high entry barriers will approach AI more slowly. Their established revenue streams make them less worried about it. 4. Critical Use-Cases: In high-stakes fields, full AI adoption will take time due to the need for trust and reliability.
-
Amit Manchanda posted thisThe more I deep dive into a particular sector, the more I see redundancy, repetitive tasks, and manual processes. Each sector would take it's own time to adopt AI (just like it did with Digitalisation)! 🏭
-
Amit Manchanda shared thisStages of ML Development in Fraud Detection Systems 🚀 🖼️ (Image generated by ChatGPT) As an ML Engineer, I've seen firsthand how these technologies evolve to outsmart fraudulent activities. Let's look at its stages: 1. Rule-Based System 📊: Begins with simple 'if-then' rules. Basic, yet foundational. 2. Machine Learning Models 💻: We introduce models trained on minimal data or pre-trained models. These assist agents by highlighting potential fraud, streamlining decision-making and making agents more effective and efficient. 3. High Precision and Low Recall Models 🎯: Focused on accuracy, these models target obvious frauds while minimizing false positives, automating a small fraction of decisions. 4. Bigger Models with More Data 📈: With more data, models detect a wider range of frauds while maintaining precision, still involving agents for complex cases. 5. Full-Automation 🤖: The goal! Advanced models handle most decisions, significantly reducing human intervention. These stages aren't just for fraud detection; they're a blueprint for ML applications in various sectors. 👇 What are your thoughts on ML's future in other areas? #MachineLearning #FraudDetection #AIRevolution #DataScience
-
Amit Manchanda shared thisWhy I am Bullish on AI 🚀 🕒 The time to adopt a technical innovation has been ever decreasing. It took a century to normalize and use the steam engine. Decades to normalize the Internet, and just one year to normalize and use ChatGPT (or other AI) everyday. 🧠 Consider the advancements over the last two years: GPT and other technologies have been crossing the state-of-the-art and human boundaries. Imagine what will happen in the months and years to come! 🌐 Accompanied by strong global interconnectivity, collaboration, and immense computing power, all the pieces are falling into place. AI is one of the biggest technology opportunities of the century, and I am not just bullish on AI; I am bullish on the speed of improvement in AI. Let me know what you think! #ArtificialIntelligence #TechInnovation #FutureTrends #AIAdoption #DigitalTransformation #ChatGPT #TechnologyLeadership
-
Amit Manchanda shared this🌟 Transitioning from an ML Engineer to an AI Product Manager, and finally, to a Business Leader has been a journey of constant evolution and learning. Here are the pivotal lessons from each phase 🚀 As a Machine Learning Engineer: 🧩 Foundational Rule-Based Systems: Most of the production ML systems have started as rule-based systems. 💡 Dynamic Business Logic Layer: Business logics are ever-changing and crucial. Thus, it's critical to have a dedicated layer managed by business professionals. 🔄 🌐 Power of Data: Prioritizing quality data over complex algorithms. 🎯 Model Explainability over Marginal Accuracy As an AI Product Manager: 🛠️ Graceful ML Failure Handling: ML models are probabilistic and not always perfect. It's crucial to have strategies in place for when models fail to avoid frustrating the end-user. 👥 User-Centric Approach: Focus on user problems, not just model accuracy. UI-UX / Engg solutions are crucial in areas of low accuracy ML. 🔄 Continuous ML Improvements (MLOps): ML products improve over time. Monitoring model performance, managing data pipelines, and automating model retraining is crucial in production. As a Business Leader: 🚀 Personalization and Automation: Top areas to leverage AI. 🌱 Building AI Culture Takes Time: Allowing teams to experiment and learn from failures is essential. The first few AI projects are often learning experiences. ⚖️ Balance Ambition with Practicality: Expecting immediate full automation from a Large Language Model (LLM) is unrealistic. AI initially suggests and assists, evolving to automate routine tasks with necessary human oversight, finally to a full-automation. 🤔 Where do you think AI is headed in the next decade? Let’s brainstorm! #aileadership #machinelearning #productmanagement #businessstrategy #techinnovation PS: The image is generated by ChatGPT, thus incorrect spelling. 😛
-
Amit Manchanda shared thisWe are looking for candidates with Ph.D. in Computer Science to join us as an assistant professor of Computer Science in Sonipat, Delhi NCR. Please reach out to me or apply at the below link. 🚀Amit Manchanda shared this"In a world driven by technology, education is the key to unlocking limitless possibilities." - Werner Vogels, CTO of Amazon Join us in revolutionizing education and technology! Newton School and Rishihood University have joined hands to drive a never seen before transformation in higher education for Indian students. We're hiring an assistant professor of computer science in Rishihood University Sonipat, Delhi NCR. If you have a Ph.D. in computer science or a related field from a top tier institute and are passionate about teaching, this is the perfect opportunity for you. Apply now and join our team! #educationrevolution #techinnovation #joinourteam #technology #teaching
-
Amit Manchanda shared thisHiring Alert 📣 📣 📣 We, at #NewtonSchool, are breaking barriers in #edtech industry. We are actively #hiring for #frontend #sde roles to achieve our vision. If you are interested or you know someone who fits the role - please reach out to me or apply at the given links: #SDE1 - https://lnkd.in/gSwAU2ZR #SDE2 - https://lnkd.in/gGNTy6c4
-
Amit Manchanda shared thisWe are looking to hire Product leaders at Newton School. If you know someone, please refer them to us!Amit Manchanda shared thisHey Zomato, we're sure we'll love your references just like we love your notifications. Refer karne ke liye link: https://bit.ly/3yUeZTA #zomato #hiring #productleaders #food
-
Amit Manchanda liked thisAmit Manchanda liked thisNot a big celebration, just a meaningful one. Every step here carries a story of effort. Grateful for the people building this every day. This is just the beginning. #NewtonSchool #NewtonSchoolLife #TeamWins #GrowthJourney #MilestoneMoment
-
Amit Manchanda liked thisAmit Manchanda liked thisSnabbit intern accidentally sent: “Test” Most brands would panic and delete it. Snabbit turned it into an opportunity 😂 Honestly these small moments feel more human than perfectly polished marketing. #Marketing #Branding #SocialMedia #Funny
-
Amit Manchanda liked thisAmit Manchanda liked thisPeer benchmarking is useless when the peer set is wrong. I keep seeing newly public companies put Microsoft in their peer group because both mention AI in their risk factors. That is the entire overlap. Different market cap, different segment count, different audit firm, ten years of filings on one side and three quarters on the other. Three failure modes keep repeating. The fix is unglamorous, and it's the part most teams skip. Most of the answers are already sitting in the right peer set. The work is finding them. #SECReporting #PeerBenchmarking
-
Amit Manchanda liked thisAmit Manchanda liked thisAman Gupta JUST COPIED MY HOMEWORK So Aman Gupta (yes, THE boAt guy) just posted about hiring interns with a fridge full of Diet Coke. Cool flex! Except... checks notes ...I did the EXACT same post 1 YEAR AGO. Same energy. Same fridge flex. Different beverage choice. The only difference? His fridge: Diet Coke 🥤 My fridge: Corona 🍺 Corona > Coke. Fight me. 🍻 Aman , if you're reading this (and I know you're not)... great minds think alike. But I was the great mind FIRST (Kidding. You are an inspiration). 😎 Also, respect for the OFF/BEAT journey. But next time you need fridge inspiration, just tag me. No need to wait a year. #StartupLife #FridgeWars #LinkedInComedy #OffBeat #BoAt #IDidItFirst #BeerVsCoke
-
Amit Manchanda liked thisAmit Manchanda liked thisI’ve been building Mine Palace (https://lnkd.in/gKrBsJFV) for the OpenCode Buildathon. The idea is simple: It turns a markdown or Obsidian vault into a Minecraft memory palace you can actually walk through. Folders become districts, notes become places in the world, and the project generates a layout plus deployment artifacts so your notes can turn into something spatial and explorable. It’s still a work in progress, but I’m excited by the direction: - a more playful way to interact with knowledge - a memory-palace-inspired interface for personal notes - a weird but useful bridge between note-taking and games More at: https://lnkd.in/gKrBsJFV PS: They wanted me to post about the ideas we are working on ._. so here is an LLM generated post. The things people have do to get engagement for their hackathon projects. #OpenCode #Buildathon #Minecraft #Obsidian #KnowledgeManagement #IndieHacking #CreativeCoding
-
Amit Manchanda liked thisAmit Manchanda liked this𝗪𝗵𝗮𝘁 𝗱𝗼 𝘆𝗼𝘂 𝘄𝗮𝗻𝘁 𝘁𝗼 𝗴𝗲𝘁 𝗱𝗼𝗻𝗲 𝘁𝗼𝗱𝗮𝘆? If updating your US-Iran risk factor is on the list this quarter, you're not alone. Every cycle: pull last quarter, research what's changed, benchmark peers, check OFAC, rewrite, chase legal and audit, reconcile redlines. Days of work. Two paragraphs of output. 𝗙𝗶𝗻𝗿𝗲𝗽 𝗗𝗿𝗮𝗳𝘁𝗶𝗻𝗴 runs the whole cycle. A dedicated AI specialist per disclosure that researches the live context, benchmarks peer filings, and drafts in your voice. 𝘋𝘳𝘢𝘧𝘵 𝘵𝘩𝘦 𝘸𝘰𝘳𝘬. 𝘛𝘳𝘢𝘤𝘬 𝘵𝘩𝘦 𝘦𝘥𝘪𝘵𝘴. 𝘖𝘸𝘯 𝘵𝘩𝘦 𝘦𝘷𝘪𝘥𝘦𝘯𝘤𝘦. Live trigger → peer-benchmarked, reviewer-ready redline. Every change sourced. 👇 What's eating your week? Curious in the comments. #SECReporting #RiskFactors #10Q #FinrepAI
-
Amit Manchanda liked thisAmit Manchanda liked this"𝗣𝗲𝗵𝗹𝗶 𝗸𝗮𝗺𝗮𝘆𝗶 𝘀𝗲 𝗶𝗣𝗵𝗼𝗻𝗲 𝗹𝗲 𝗹𝗶𝘆𝗮 𝘆𝗮𝗮𝗿..."🧿 A few days ago, I celebrated my birthday and officially completed my teenage years 🥹. When I look back, what I have achieved in the last 2 years after joining college is huge for me. I never thought I'd be earning during my first year of college, but thanks to a summer internship and some cool freelancing projects, it helped me earn some pennies. Being able to self-fund some of the gadgets I use is a feeling I can’t quite put into words. But honestly, this post isn’t about the gadgets. It’s about what it took to get here. It represents the reality of the past year balancing my 𝟮𝗻𝗱-𝘆𝗲𝗮𝗿 university academics with late-night coding sessions. It is the result of debugging MERN stack applications 🥹 (after the mess created by AI), writing endless Python scripts, diving deep into building AI models... and 𝙡𝙤𝙤𝙤𝙤𝙤𝙤𝙤𝙣𝙜 meetings to build something that I can call a startup. But the biggest contributors to this journey are my parents. None of this would have been possible without their sacrifices and the way they constantly keep me motivated. My college Newton School of Technology X Rishihood University, also played a huge role in this. Knowingly or unknowingly, those long-hour sessions on internships and entrepreneurship helped me a lot. #FirstIncome #Milestone #TechJourney #Gratitude #CareerGrowth #SoftwareEngineering #internships #internship #python #genai
-
Amit Manchanda liked thisAmit Manchanda liked thisLast Thursday, we launched 𝐀𝐈 𝐚𝐠𝐞𝐧𝐭𝐬 𝐛𝐮𝐢𝐥𝐭 𝐟𝐨𝐫 𝐜𝐨𝐧𝐬𝐭𝐫𝐮𝐜𝐭𝐢𝐨𝐧. It was a big moment for us and for the industry. We were expecting around 150 participants to join us. 230+ walked in. But what stayed with me most wasn’t just the turnout. It was what happened in the days that followed. Within a week, we started seeing inbound conversations from some of the most respected engineering & construction firms in the country further adding to the growing revenue pipeline. The response tells us something important. The industry is not curious about AI anymore. It’s ready to use it. Going back to the launch day, I clearly remember - the moment the product went live, the room transformed - conversations started at every table. Leaders were discussing real use cases, real project risks, and how technology could directly improve outcomes on site. The inquisitiveness was practical, grounded, and immediate. And the fireside discussion exactly captured why this shift matters. Madhu Aravind, Sr. Manager Projects, ITC Limited shared a perspective that resonated strongly: for organizations to truly leverage AI, the data foundation is the key. ITC has been investing in that discipline for years, and it highlights why vertical SaaS platforms are essential for scaling companies. Parallely, Subhakar Rao Surapaneni, Chairman Champions Group rightly pointed out during the conversation: the stakes in construction are so high that even a small mistake in a BOQ can lead to crores in losses. That reality defines this industry. Precision is not optional. Decisions carry weight. What also made me proud that day was our team - at some point, roles disappeared. Sales, product, customer success, engineering - everyone stepped in wherever they could help. Explaining, discussing, and working alongside customers to make the product immediately useful. That level of ownership is part of how Powerplayers operate every day. It felt like the start of a new way of working for construction. Grateful to everyone who showed up, engaged deeply, and is helping move this industry forward. Powerplay Shubham Goyal #PowerplayAI #ConstructionTech
-
Amit Manchanda liked thisAmit Manchanda liked thisA few weeks ago, Newton School of Technology students travelled to China. This week, that journey came full circle. We had the opportunity to host Mr. Qin Jie, Consul General of the People’s Republic of China in Mumbai, who met the students who had been part of the China immersion, along with NST faculty. In the interaction, he shared how encouraging it is to see young Indian talent engaging with China not just through technology, but also through culture and shared experiences. As Consul General, Mr. Qin Jie represents China’s diplomatic and economic engagement in one of India’s key regions, making conversations like these meaningful beyond the room. The students’ China visit was also covered by Chinese media, showing how closely such exchanges are being followed. (Link in comments) What started as a student immersion is now leading to direct dialogue between ecosystems. That is how bridges get built 🇮🇳 🇨🇳 . Siddharth Maheshwari Himalay Reddy Krushn Dayshmookh Deepak Kumar Gour Prayog Mali Saurabh Kumar Faiz Mehdi Gaurav Gehlot
Experience & Education
View Amit’s full experience
See their title, tenure and more.
or
Projects
-
Digitalization of Scooty
Aimed to retrofit a normal Scooty to a dockless Scooty to be picked and dropped anywhere by anyone, keeping the security of the scooty intact
Involved keyless engine switching, opening and closing of handle and seat lock, digitalization of fuel measurement and GPS tracking -
Dynamic Speed Limit
Proposed a novel algorithm to reduce congestion in a city by determining the road specific speed limit at a particular time, and verified the results using simulation.
Algorithm took into account traffic in the city and the situational factors of a specific road like bicycles, pedestrians, and heavy vehicles to calculate the speed limitOther creators -
-
ALUM - A LUMPED UP MARKET
Developed an Android based zero cost and highly localized advertising solution for small businesses
Involved developing a buyer side application to remind him of shopping in a nearby store in GPS free and internet free way
-
Energy generation using Piezoelectric Effect
This project was presented in Make In India Hackathon, IIT Bombay. Tiles were made using piezoelectric material which would generate energy when pressed. The motive is to place such kind of tiles in public places that experience enormous footfall and consume the energy produced in some low power devices there itself.
Other creators -
-
SLAMS (Self Learning Automated Monitoring System)
The project was done under aegis of electronics section, IITR and had been presented in Inter IIT Tech Meet-2016 held in IIT Mandi. This project involved creating an android and a web portal to control various devices in home. The app would also maintain a log of family members entering and exiting the home. The system also had measures for door security and automatic curtains based on light intensity outside the room. Artificial intelligence (AI) was implemented to control the fan speed.
Other creators -
-
Identified Flying Object
A windows application was developed during the 24-hour Code.Fun.Do hackathon organized by Microsoft. The application was made to control the movements of Parrot AR Drone 2.0 with the gestures that person made using Kinect 2.0.
Other creators -
-
Your Society Needs You
We made an application to reduce the crimes against Women. In case of emergency, the application took the data of the situation and location via GPS. This data was sent to an Emergency Helpline so that help can reach to the one in need as soon as possible.
Other creators -
-
Movie ticket booking system
A movie ticket booking platform developed in Java that allows users to watch trailers of upcoming movies and purchase tickets.
Other creators -
-
Generative Adversarial Text to Image Synthesis
-
• Worked on the Automatic synthesis of realistic images from detailed text descriptions as Bachelor's Thesis Project
• Generated a comparative result on birds and flowers involving Conditional GAN, Conditional WGAN, Attention GAN
• Used skip-thought vectors and RNN text encoding for text vectorization
Other creators -
-
Comparison of image processing techniques on Retinal Blood Vessel Segmentation
-
• Automatic segmentation of blood vessels in fundus images is a binary classification problem of classifying a pixel as vessel or non-vessel.
• GPU implementation of max-pooling CNNs was used for segmenting purpose. The model was tested on various inputs such as pixel intensity, FFT coefficients and 2D Gabor wavelet responses of the images.Other creators -
-
Hinglish Date-Time Parser
-
See project• Mentored the project which aimed at creating an algorithm to extract the date and time-related words from Hinglish input and to parse them. Hinglish input consists of Devnagri script but written in Roman Language.
• For Example, if the input is "Kal shaam ko paanch baj kar 18 mins 20 seconds main meri meeting hai." then algo output is Date -> [30/07/2017] Time -> [5:18:20 PM]. This information has many useful applications. -
Smarter Walking Cane
-
See project• Mentored the project which was presented in Srishti 2017.
• A cane for aiding blind people via obstacle detection and scene description. Used ultrasonic sensor for obstacle detection. The cane includes a camera that captures an image at the push of a button and then describes it in 2 line. -
Smarter Phone
-
• An Activity based Smart health assistant that can provide health tips, product recommendations etc by analyzing daily activities such as walking, sitting, standing etc. Developed using Smartphone's accelerometer sensor, Machine learning (CNN). Currently developed for Android.
Other creators -
-
A systolic architecture for face recognition system based on principle component neural network
-
• An efficient self-configurable systolic architecture for very large scale integration implementation of a face recognition system.
• Implemented the Hebbian Learning algorithm on XILINX Spartan-3E.
• Able to identify 14 out of 15 different faces which were not used while training of datasets.Other creators -
Honors & Awards
-
3rd prize in Garage AI Hackathon
Microsoft
-
2nd prize in WiFi Th!nkFest
Mojo Networks
Summer Wi-Think Fest was an attempt to harness the imagination of young minds for extending the power of WiFi beyond just a means of Internet access and encourage students to take an out-of-the-box approach towards developing new applications of WiFi technology.
-
World finalist in Valeo Innovation Challenge
Valeo
Selected as one of the 7 world finalist teams of the Valeo Innovation Challenge 2017.
-
Digimize'17 National Winners
Hitachi Consulting Software Services( India)
Digimize, a national level competition focused on technology optimization, was organized by Hitachi Consulting Software Services India.
-
Srishti 2016
IIT Roorkee
Secured 1st position (IoT Category) in Srishti 2016, the annual intra-collegiate festival of IIT Roorkee
-
Finalist
Make In India Hackathon, IIT Bombay
Selected among top 10 teams from all over India in Energy theme for the finals of Make in India Hackathon. We were selected to present the prototype developed for Energy Harvesting using Piezoelectric Effect.
-
Inter-IIT Tech Meet 2016
Indian Institute of Technology (IIT)
Represented IIT Roorkee and bagged fourth rank in 4th Inter IIT tech meet held at IIT Mandi in the year 2016
-
Best Application Award
Sankalp, IIT Roorkee
Winner of the inter-college competition during Sankalp 2015 (Annual Social Fest, IIT Roorkee) for an application on Women Security.
-
Merit Certificate in Chemistry in AISSCE 2014
Central Board of Secondary Education
For scoring the highest marks in Chemistry in AISSCE Class XII Board Exams 2014
Other similar profiles
Explore more posts
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