Ankith Harathi

  • Student

Activity Feed

On November 15, 2018, Ankith Harathi commented on 3D Printed Cars – The Future of Auto Manufacturing or a Pipe Dream? :

Love the article and this topic. 3D printed auto parts also introduce the possibility of efficiency and performance gains that would never be possible in their absence. Koenigsegg, a Swedish hypercar manufacturer, 3D prints its turbo chargers on its top of the line Agera. The 3D printed turbos have a variable A/R ratio meaning that the inner geometry of the turbocharger spiral actually changes depending on the engine RPM, maximizing power producing efficiency – something that was incredibly technically challenging and added a tremendous amount of weight previously. This type of turbocharger is now being developed for smaller, < 1L engine cars, to help produce more power per liter and make economy cars more efficient. There's a LOT more that can be done!

Great read! I do think AECOM should be concerned about regulation, and also the perspective of insurance companies on 3D printed houses/office buildings. Similar to autonomous vehicles, I think insurance can pose a similar problem: would companies be willing to cover houses that are 3D printed? Who is liable if the house collapses or has any other structural damage – would the 3D printing company themselves have to share in the cost? Lots left to be answered in this space but very exciting to hear about its development!

On November 14, 2018, Ankith Harathi commented on How crowdsourcing is changing the Waze we drive :

Awesome article! I wonder if Waze has faced any repurcussions regarding its use of crowd-sourcing traffic data. In my experience with Waze, having to enter when there’s a pothole on the road, police are nearby, or there is significant traffic can distract from the actual experience of driving. Furthermore, I’ve even been shown ads while driving that I’ve attempted to swipe away or get rid of. I understand they are attempting to get very granular data for traffic; however the trade-off is driver attentiveness and safety. I wonder if there is a way for them to more heavily rely on the passive crowdsourced data and their network of map-editors rather than active data from drivers.

On November 14, 2018, Ankith Harathi commented on How crowdsourcing is changing the Waze we drive :

Awesome article! I wonder if Waze has face any repurcussions regarding its use of crowd-sourcing traffic data. In my experience with Waze, having to enter when there’s a pothole on the road, police are nearby, or there is significant traffic can someones distract from the actual experience of driving. Furthermore, I’ve even been shown ads while driving that I’ve attempted to swipe away or get rid of. I understand they are attempting to get very granular data for traffic; however the trade-off is driver attentiveness and safety. I wonder if there is a way for them to more heavily rely on the passive crowdsourced data and their network of map-editors rather than active data from drivers.

On November 14, 2018, Ankith Harathi commented on HTC Vive: Overcoming Developer Hesitance Through Open Innovation :

This reminds me of when Apple introduced the App Store – it relied extremely heavily on many smaller independent developers creating apps that iPhone users could use. Granted, Apple had relationships with large companies so they could get their apps on platform first, a large part of what drove the App Store’s success, and later the Google Play Store, was the variety and sheer number of applications available. The lack of apps was an important factor in the failure of the Windows Phone OS and Fire Phone, among other things. Along with raising a fund for developers, I think HTC and Valve should fees and any other monetary barrier that developers face in order to ensure highest chances of success for their VR platform.

On November 14, 2018, Ankith Harathi commented on Slack’s Battle Against Information Overload :

Great article! There have been a lot of recent developments to text summarization in machine learning and NLP algorithms, specifically in extraction and abstraction based summarization. Perhaps Slack could use this instead of highlights to summarize the missed messages per channel between when a user has last interacted with slack. Agree with the comment above that if people solely rely to get their information, then could be dangerous if this algorithm doesn’t work 100% of the time. One thing to accommodate that case is to include a slack specific tag, perhaps the *bold* tag so that any text delivered in bold is deemed extremely important and automatically included in all summarizations.

I think another interesting way Pinterest can leverage their massive amount of data would be to sell brands aggregated (not individual) user data on trends they’ve identified. As an example in fashion, if Pinterest notices lots of users of a certain demographic pinning images with a new clothing trend, it can offer advice to fashion brands segmentation breakdowns of items in that trend that are most popular/sought after by their users, potentially influencing the products that brand produces.