Ask HN: What problem in your industry is a potential startup?
A thread like this was posted 2 years ago. Maybe those problems have been solved or new problems have appeared. So what problem currently exists in your industry that can be taken up by a startup?
43 comments
Anyone that can make hiring better would make a fortune. Even if you can make something super specialized like hiring for ai programmers or biochemists would make a fortune.
It amazes me how many hiring platforms/tools there are. Greenhouse, lever etc. It's like I see a new one every day.
what kind of problems are you facing with hiring?
some issues:
- massive information asymmetry b/n company and applicants.
- companies are different. why conform to standard "job types" like "data scientist", "software engineer" etc when a potentially custom tailored description would do a much better job?
- In technical fields the "non-technical recruiter as the gate keeper" model breaks down.
- There are many issues if you really dig in.
Is it finding people or background checks or onboarding? What's the source of pain?
3D modeling objects (starting with retail products) for rendering on the web and in AR/VR.
The AR/VR/WebGL world totally depends on 3D content. 99% of the companies who want to put 3D content out, don't have any in 3D or if they do it's inappropriate for the medium - for example they only have 200MB objects used for photorealistic rendering.
Having good lightweight 3D models of objects/products etc... is probably a multi-billion industry itself.
The AR/VR/WebGL world totally depends on 3D content. 99% of the companies who want to put 3D content out, don't have any in 3D or if they do it's inappropriate for the medium - for example they only have 200MB objects used for photorealistic rendering.
Having good lightweight 3D models of objects/products etc... is probably a multi-billion industry itself.
I have been trying to get that service off the ground. I got the low cost 3d scanning and rendering down. Just don't have the space or funds to actually build up the library.
GetKipple.com
I don't think it's going to be done through scanning. You need more variability for texture than can be given through scanning and fixing the textures later is too time intensive.
See my response here
https://gist.githubusercontent.com/anonymous/cba53d636ac477d...
I should add reasons there is a need for change:
Vendors in the current space are "behind" in terms of technological offering. The APIs provided by vendors are often cludgy to non-existent. The UI for these applications are non-intuitive and usually an afterthought.
The top-tier vendors in this space regularly charge upwards of $5000 or more for a concurrent license
Vendors and their partners regularly charge somewhere between $1400 - $1800 per day for consulting services
Projects for LIMS implementation typically last over 6 months and many last over 1 or 2 years.
Maintenance costs typically are in the %18-20% range.
While those numbers may be typical for enterprise software, the value doesn't warrant it. Product quality is usually poor due to poor product testing. Many of the top-tier vendors have highly-complex solutions that are the result of piling new technology on top of old-messy data models year after year. A new, serious player in this space would have the advantage of starting "from scratch".
There are many benefits and reason as to why we should be seeing innovation, but we just aren't seeing it.
https://gist.githubusercontent.com/anonymous/cba53d636ac477d...
I should add reasons there is a need for change:
Vendors in the current space are "behind" in terms of technological offering. The APIs provided by vendors are often cludgy to non-existent. The UI for these applications are non-intuitive and usually an afterthought.
The top-tier vendors in this space regularly charge upwards of $5000 or more for a concurrent license
Vendors and their partners regularly charge somewhere between $1400 - $1800 per day for consulting services
Projects for LIMS implementation typically last over 6 months and many last over 1 or 2 years.
Maintenance costs typically are in the %18-20% range.
While those numbers may be typical for enterprise software, the value doesn't warrant it. Product quality is usually poor due to poor product testing. Many of the top-tier vendors have highly-complex solutions that are the result of piling new technology on top of old-messy data models year after year. A new, serious player in this space would have the advantage of starting "from scratch".
There are many benefits and reason as to why we should be seeing innovation, but we just aren't seeing it.
This looks a lot like what you're describing, are you familiar with them?
https://www.coreinformatics.com/lims/
https://www.coreinformatics.com/lims/
PLC Programming is years behind when it comes to software engineering, developing, testing and speed of deployment.
A system that would allow for Change related Issue-tracking on the CAD-files of the controlled machine. A language that would enforce component-class interfaces, thus preventing information duplication. A proofer that validates the resulting DEAs for correctness.
A system that would allow for Change related Issue-tracking on the CAD-files of the controlled machine. A language that would enforce component-class interfaces, thus preventing information duplication. A proofer that validates the resulting DEAs for correctness.
In similar vein a tool that would let you 'play' with hardware functionality/config software without needing to buy it 'simulation on a PC or similar'.
Just having a sane place that list specs for competing products or products that have overlapping functionality would be a help. Sure simple components are standardized by when you start getting to more complex things like protective relays, PLCs, or even breakers it's like companies purposefully fling specs and functionally across 5 sites and 3 PDFs to waste so much time that you don't have time to look at competitors products.
Just having a sane place that list specs for competing products or products that have overlapping functionality would be a help. Sure simple components are standardized by when you start getting to more complex things like protective relays, PLCs, or even breakers it's like companies purposefully fling specs and functionally across 5 sites and 3 PDFs to waste so much time that you don't have time to look at competitors products.
If I consider my industry going to bars to drink then the problem is lines.
In 2017, I shouldn't walk/uber 20min in 20 degree weather to a bar only to find a 30min line which also has a surprise cover charge at the door.
I've thought about this a lot and have some ideas...
In 2017, I shouldn't walk/uber 20min in 20 degree weather to a bar only to find a 30min line which also has a surprise cover charge at the door.
I've thought about this a lot and have some ideas...
I would approach this problem by opening a bar, not a startup.
Yeah, seems like there's a need for more bars instead of a complicated solution to a simple problem.
I wouldn't agree with that. Opening a bar is easy, but there is always a lot of empty bars. You need to make a popular, cool bar with a nice atmosphere that is on the perfect border of "going out" and "chilling", with a good gender ratio and you have to make it pretty popular. But that doesn't cut the lines in half in the other bars. I'm thinking there might be a solution in having a platform that allows you subscription-style entry into bars. Like pay 50$ a month and you get instant entry + 1 friend into a place. Add another friend for n$ where n is the average price of entry or determined by the hosting bar/club.
complicated how? Is it "easier" to open physical bars with usually thin margins and massive capital investment + branding + staffing up + etc...? When all is said and done, a tech solution might be easier/better.
A tech solution for what actually? The problem is that there are few (good?) bars in the area. The solution for that problem would be to open new bars.
You may disagree but I feel we are in an age of transparency. TripAdvisor, yelp, cargurus, Amazon reviews etc. Consumers want to know as much detail as possible prior to a purchase or visiting a hotel.
My big point is that one of the remaining non transparent industries is nightlife. Bars create fake lines and charge bogus covers.
I just feel like technology could improve this.
My big point is that one of the remaining non transparent industries is nightlife. Bars create fake lines and charge bogus covers.
I just feel like technology could improve this.
And they became profitable because they managed to provide a visibility service for millions of local establishments, not a bunch of crooked bars.
Heres what I'm braining: Have everyone meet up, and a driver delivers a 6 pack to location.
Its not directly the answer of the question but relevant. I believe unaddressed/overlooked problems are the greatest origin of a potential startups but not what we think an legitimate idea, actually is. For the 1st time every idea seems awesome, take time revisit after a few days, you will wonder that most of the sparks are gone by now. If it still looks promising start doing your research. This is another eliminator of an unique idea. Though it doesn't have to be unique. But you should bring your own flavor into it.
In the future buildings and factories should be constantly monitored and controlled from the cloud. But most of these assets are currently controlled from on-premise control boxes/computers called BMS systems. If you start a company to integrate the data from these systems (many manufacturers, many formats, many proprietary) and pipe them to the cloud and put a unified control scheme on top, you could potentially do really well.
One thing I have noticed is interaction during car servicing. When giving my car in for service, the person in charge at the car company usually has to call me on the phone and tell me to pick up the car. Perhaps this could be automated where the mechanic presses a button on his/her end and the car owner automatically gets a notification (via app?) to pick up the car.
An SMS notification would do just fine, right?
This seems like a small feature to add to their invoicing app, not a business.
Also, there's quite a number of car servicing startups now, like YourMechanic.
Also, there's quite a number of car servicing startups now, like YourMechanic.
My local Lexus dealership does exactly this. The software platform they use will send me a text message when my car is ready.
Issuing currency and making laws.
I'd love to tell you but I'm working on it already.
Never be afraid to share your idea with the public. People don't care enough to steal it from you, and those that would steal it wont have the industry knowledge or contacts to launch it successfully.
See my blog post on using strangers to vet ideas: http://jeremyaboyd.com/validating-your-ideas-on-strangers/
See my blog post on using strangers to vet ideas: http://jeremyaboyd.com/validating-your-ideas-on-strangers/
> Never be afraid to share your idea with the public. People don't care enough to steal it from you, and those that would steal it wont have the industry knowledge or contacts to launch it successfully
...sometimes
http://www.cnbc.com/2017/01/30/a-24-year-old-made-345000-by-...
...sometimes
http://www.cnbc.com/2017/01/30/a-24-year-old-made-345000-by-...
Automating retro-sysnthetic analysis using AI.
A robot that can hold my baby's bottle.
Well it's not a robot but it will hold your baby's bottle:
https://www.thebeebo.com
https://www.thebeebo.com
I've been pretty successful using a blanket :).
Nursing pillows are even better, as they give enough space for the laptop to fit :P.
http://www.toysrus.com/graphics/product_images/pTRU1-2302555...
Nursing pillows are even better, as they give enough space for the laptop to fit :P.
http://www.toysrus.com/graphics/product_images/pTRU1-2302555...
A high end upwork or elance that only has semi vetted developers, e.g those with github commits > 100.
A simpler and cheaper version of meetedgar.com that keeps retweeting my blog posts on auto pilot.
A simpler and cheaper version of meetedgar.com that keeps retweeting my blog posts on auto pilot.
have you used toptal before?
Gigster and toptal a) have their own vetting mechanism that misses out on lots of good developers, b) charge a premium for that which is unnecessary, and c) are a service not a platform.
What I want is a lightweight platform where any github dev can showcase their code and say hire me for x. I don't want any third party vetting. Github itself could do this. We need a better freelance marketplace.
What I want is a lightweight platform where any github dev can showcase their code and say hire me for x. I don't want any third party vetting. Github itself could do this. We need a better freelance marketplace.
Toptal, gigster.
What is the pricing at meetedgar.com?
And what features they have that are essential and what are disposable?
Their site seems to be purposefully opaque regarding these points.
Their site seems to be purposefully opaque regarding these points.
Starts at $50 per month. You need to setup a "library" of blog posts (yours and others) and it will keep retweeting them on autopilot. I want something for $10 per month that simply connects to my blog and retweets posts all day. I'd imagine many bloggers would want it as well.
I was thinking about that these days. If I build it, will you be my first customer?