Machine learning, LLMs, recommendations and applied AI.
In a world where every delivery is 2 minutes away, and every service is only a tap away, the need for devices to be smarter, faster, and more secure constantly is high. At the same time, users are more aware of their data privacy and expect seamless experiences without the lag…
By — Kshitiz Swaroop ( Data Scientist, Data) In a world overflowing with data, finding correlations is easy. The real challenge is understanding causality. This article presents 2 case studies where the methodology was used to improve the organization wide estimates on the…
This blog post explores an effective strategy to streamline fragment management in your app by revising the feed setup and implementing lazy loading.
Apple has just unleashed some serious game-changers with iOS 14 and the iPhone 14 Pro. Picture this: a slick Live Activity widget to spice up your lock screen, and hold onto your hats because Dynamic Island has landed, folks!
By — Resham Wadhwa ( Data Scientist, Data) At Urban Company, we act as a matchmaker between customers and service professionals for home services like salons and cleaning. Our goal is to find the right professional, ensure they accept the request, arrive on time, and deliver…
In this post we will discuss the evolving architecture for Sharechat’s event streaming platform. We will deep dive into what was our first generation architecture, what problem statements we constantly faced and there-fore strived to solve.
In the fast-paced world of technology, orchestrating workflows efficiently has never been more crucial! That's where Temporal comes into play.
Boost your productivity with Hasura GraphQL and Portkey This blog illustrates how Hasura and Portkey can boost your productivity and help you and your company stay competitive.
To leverage LLMs and build an SRE assistant, we must augment the business context (of being an SRE) to LLMs by prompt engineering. ClickHouse is a database designed to allow us to fetch the relevant context for a query, and Hasura acts as a secure data-access layer.
Implementing more accurate manual testing and automated integration tests has significantly boosted our developer velocity.
As part of Hasura’s evolution, today we’re introducing Hasura Notebooks, a tool designed to facilitate the swift prototyping of cutting-edge GenAI applications.
From hearts to cakes, to a cool pair of sunglasses, a MOJ user has many choices to choose from when it comes to virtual gifting. But what else has virtual gifting enabled?
You are trying to create a Responsive Card component for a design system for your website. Here's a blog where we detail container queries, and how to do it in Tailwind
Google Image Picker is an npm plugin that helps you upload images directly from your web application without going through all the hassle of searching on your browser, downloading, and then uploading to the server. Here's how!
At ShareChat and Moj, we send around eight billion notifications per day. In this blog, we detail how we drive maximum efficiency, without hampering the user experience
While building apps with react, you would have come across bunch of redundant lines of code so that your state understands you're doing something with it. Here's how to use proxies to make React state mutable, & stop destructing your store
We, at ShareChat, believe in working smart. Here are some tips and tools to ensure you are at your productive best as a developer
Using Custom React Hook, we explain how to avoid memory leaks in the components, make Atomic API transactions with respect to your Component, and less number of API calls altogether
A smooth experience is paramount to any app and cannot be compromised. In this blog, we explain how we used React swipeable screens in Jeet11 to solve a number of issues
While developing a user-facing application, any developer aims to provide the smoothest possible experience. In this blog, we detail, how a frame is displayed, general issues with smooth scrolling, transitions, or animations in iOS
ShareChat’s acquisition of Mx TakaTak required their data to be transferred from their Cloud infrastructure to ours in the shortest possible time to bring both universes together. Here's how we did it
CORS errors can be a pain to front-end developers while developing. In this blog, we explain in detail the fixes that can help in speeding up the development process without getting blocked by these errors
One of the demands that we’ve frequently received was for animation in the form of flying hearts and flying emojis/reactions for live-streaming on the Moj app. Here's a detailed look at how we built an in-house solution
3MASSIV: Multilingual, Multimodal and Multi-Aspect Dataset of Social Media Short Videos, which was presented at the Conference of Computer Vision and Pattern Recognition (CVPR), 2022 held at New Orleans, Louisiana
In this blog, we deep dive into our infrastructure journey talking about the system improvements we have made, the rationale behind them & what impact it has had so far.
Being a social media company we maintain a lot of counters; from views and likes to many more internal ones related to revenue spending etc. Updating a count might seem simple but as always the real problems come with scale.
Short videos are often part of our web-cards, PWAs etc. With our compression service, those videos will be ready-made for uploading without any extra work.
🤔 Ever wondered how many times a developer builds a project and how much developer time gets wasted on this? We consider build time as a break to grab a coffee or chat with someone or just start browsing the trending news.
Modern websites demand speed. As web developers we have to work not only on javascript but provide the best user experience as well.
Ever wondered how you are able to find relevant and entertaining content every time you go to the Moj or ShareChat “Home” page? Machine Learning is at the core of this complex machinery!
Advertisers want to run their advertisement campaigns on ShareChat and reach their target audiences. Our aim is to show the advertisement to the most relevant audience so that the advertisers' goals are met.
A few months ago, we at ShareChat decided to go on a documentation coverage spree. Not long into the journey we realised the resources for measuring documentation were limited and would not meet our expectations.
We wanted to give our players the ability to generate the best declaration for their cards. Here's how dynamic programming got the work done.
ShareChat AI team made its presence felt at the recently concluded EMNLP Conference 2021 held in the Dominican Republic. Here’s what we spoke about!
In this approach, an additional calibration step is involved, wherein a representative dataset is used to estimate the range of activations using the variations in the dataset.
We wanted to make the whole gifting experience feel very fast but this was not possible if we kept waiting for the API response.
Though our problem from an outside perspective is similar to OCR in the wild problem, the primary distinction is content intelligence.
While designing the OCR system, we had both computational complexity and correctness in mind, as the system needs to scale up to the heavy usage and use-cases. We tried models, like Single Shot Detection[4] and R2CNN[5],
Majority of existing work tackles only one of the aforementioned problems at a time. We at ShareChat and Moj, on the other hand, are working towards a robust all-in-one framework to considerably improve the quality of the UGC,
Since our use cases deal primarily with spoken audio, we use the popular overlap and shift paradigm used to analyse speech signals with a 25ms window length and 10ms shift.
For each language and content type, we have a different configuration for extracting features as well as separate Milvus clusters. This allows us to tackle challenges that come with each language/content-type
A PID controller continuously calculates an error value as the difference between the desired value and the measured value and applies a correction factor so that the measured value becomes equal to the desired value.
The jobs are written in a way that most of the heavy lifting in terms of computation is done by BigQuery itself (parsed from GKE metering). We simply aggregate the results and push those back to another BigQuery dataset.
Millions of posts are created every day on ShareChat and Moj which makes timely and precise detection of Integrity Violating Content (IVC) crucial for ensuring the safety of our users and
If you have troubleshot Distributed Systems, you would realise that the maximum time is spent in trying to find out where the problem emanates from. Once the system-at-fault has been identified, the fix itself becomes very straightforward.
We used philosophies of open tracing for extracting various metrics from every single resource and collected Externally to create actionable items and build more intelligent infrastructure.
While our models demonstrate amazing speed and accuracy, they are data-hungry and require many labelled data for training. Labelling this data is a time consuming and costly process as it requires manual inspection.
Videos are an expressive, engaging and entertaining form of content. However, from a machine learning point of view, videos are incredibly complex as the information flows in both spatial and temporal dimensions.