HN2new | past | comments | ask | show | jobs | submit | martinshen's commentslogin

Who do you think backs those Clearco loans?


My 2 cents... you need to start doing cold sales and it'll be very difficult as you'll get flat-out rejected 95/100 times.

Learning about sales will feel more productive and in your comfort zone but you should start by going out there and talking to customers. Get out of your comfort zone.

I'd start by going door-to-door so you can start gathering feedback from SMBs and get a sense of the true ICP. Once you've closed ~10 clients this way then you should consider using a sales engagement platform like Outreach, Hubspot or Salesloft which will automate the cold email to call. If it's a 1-call close type sale, then you can use something like https://www.mojosells.com/. You can buy lists from Zoominfo or otherwise.

That all said, any type of human-centered sales motion in North America will require a minimum annual contract value of $10,000/yr to scale up.


Zippin | https://www.getzippin.com/ | Toronto

At Zippin we plan to banish standing in line—for good—with our checkout-free technology that’s easy and cost-effective for retailers to deploy, and greatly improves customer experiences in-store. What’s more, Zippin offers unparalleled inventory tracking and insights to ensure the right products are in the right place, at the right time.

Our new Toronto office is looking for a new Business Development Manager with experience with enterprise sales that requires significant integration.

Email me at martin@getzippin.com


Have you tried using object attributes/features to enable tracking instead of IOU-based tracking? I suspect this tracker falls apart when there are significant obstructions. Thoughts on a Recurrent YOLO or DeepSORT based tracker?


Yes. I am actually looking into this idea. I thought instead of recurrent yolo, Correlation Filter is much simpler.

I was referring the following papers:

1. https://ethz.ch/content/dam/ethz/special-interest/baug/igp/p...

2. https://zpascal.net/cvpr2018/He_A_Twofold_Siamese_CVPR_2018_...

3. http://openaccess.thecvf.com/content_ECCV_2018/papers/Yunhua...


I've found that IOU based tracking is surprisingly reliable (my dataset was primarily surveillance footage with a static camera) if you take measures to smooth the data and predict object motion using a kalman filter.


Kalman filter is also one good way to reliably keep track of objects under significant obstruction.


Why would they pay you for your data? How much is it worth?

If you click on 0 ads per year, you generate $0 for Facebook and therefore your data is worth $0.

For a personally targeted (using your data) ad click on FB, you're looking at ~$3.25. For a web targeted ad click on a niche website, you're probably looking $0.50-$2.50.

In other words, your "data" is worth $0.75-$2.75 per click of revenue

FB spends 17% of revenue on direct costs and another 18% on marketing. Let's ignore R&D improvements and assume 35% of costs to serve that revenue.

Profit per click for FB = $0.49-$1.79 per click.

Please consider that these clicks must have real intent behind them. In other words, there should be at least 10% chance you'd actually buy that product.

In other words, your data is not worth that much money.

...

On the flip side, FB should offer a paid ad-free version of their services but who honestly would pay $10/mo for FB + Instagram etc.?


> On the flip side, FB should offer a paid ad-free version of their services but who honestly would pay $10/mo for FB + Instagram etc.?

IIRC they don't make anywhere near $10/user/month. If you gave me the option to pay, say, $25/year for ad-free, tracking-free, and it was implemented in some way that I could trust, then I might be willing to participate with FB again. I'm hoping, actually, that someone will come along with exactly this business model so that we can have social networking for friends and family without the dark patterns. Needs to get everyone on board, of course, which is the hard part.


There's an adverse selection issue. Let's say you make $10/month/user from ads, and offer the option to pay $10/month to not have ads. Users with less money are less likely to take you up on that, and are also less valuable as an advertising audience, so you start bringing in much less per month per ads-only user.

Advertising effectively allows a company to charge richer users more, which you lose when you switch to a flat fee.


That is an angle I had not considered, thanks!

It seems like a chicken & egg problem, in some ways. A business that was built more traditionally than a unicorn could still make good money at $10/user/month, certainly it's possible to build Facebook-level social networking without employing nearly as many developers as they do. But without the network effect, the value is not there. I don't want to pay to be by myself.

Maybe it could work as a SaaS, selling distinct social FB-style closed social networks to families, with a future option to interconnect those.

Probably someone has already tried this and failed. I'm not an idea guy, just a coder ;-)


I think app.net (https://en.wikipedia.org/wiki/App.net) tried that approach and failed to get traction.

I view it kind of like Craigslist. If I put something there for free tons of people are interested. If I charge even a dollar the interest drops by a magnitude.


Some ex FB employees did make this. It was on techcrunch in the past 1 year, I think. Just found it, it shut down.

https://techcrunch.com/2018/07/30/fabric-offers-an-alternati...

https://medium.com/@fabric.me/sunsetting-the-fabric-app-deb2...



Well, that's better than I remember :-). Or I'm thinking worldwide average. I'd probably still do it if the software was good enough and the network complete enough.


I agree. The whole point is that they are already providing a service to you in exchange for your data—a service that is quite expensive to run. If you want to sell somebody your data, that's fine. Pick how you want to be paid. Services or cash.


This sort of hits a key issue for me - my data is pretty worthless to them, but it's quite valuable to me.

Why don't we realize that privacy is a resource and the current consumption of that resource to drive advertising is incredibly inefficient.


Your data is worth whatever someone will pay. It isn't just about ad revenue, it's about the market research that becomes a behavioral futures market. This app gives you a personal data value estimate (average it's a couple hundred dollars/year): https://app.fastgarden.io/assessment


That is how much the market values data, but markets are not the ultimate decider of worth. Additionally, different participants in markets place different values on the goods being traded, otherwise trade would be useless - the participants consuming our privacy right now value that privacy incredibly cheaply.


Passenger AI | ML & CV Engineer | Toronto, Onsite | $110-150K + 0.2-1.0% equity

Passenger AI builds remote monitoring tools for autonomous taxis and shuttles. We’re looking for a creative machine learning engineer ideally with computer vision experience. You would work with our 6-man engineering team and data ops to build models and be able to test them immediately in the real world.

Email me (CEO) martin@passenger.ai and I'll respond promptly. No recruiters or recruiting services.


Passenger AI | Toronto | $100-160K + 0.1-1.0% equity | Full-time Onsite

I’m the founder/CEO. In 2017 I moved from startups in SF to Volkswagen in Detroit to learn about the massive industry we know will transform over the next decade. I left this past April to start Passenger AI which is building remote monitoring tools for robotaxi operations. In the past 8 months, we’ve raised from top-tier VCs, built up a team of 8 (Mozilla, Lyft, Pivotal etc.) and brought on some great customers.

We moved into a new office last month with a garage so we can touch/feel our product everyday too!

We’re hiring engineers for DevOps, Media Streaming and ML/CV. (https://angel.co/passenger-ai/jobs)

Email me at martin@passenger.ai if you want to learn more.

No recruitment-type services please.


By far the biggest problem in Toronto is that we do not have a deep startup tech talent pool. They all leave for the SFBay.

Why? Startups here pay engineers 1/3 to half of what they'd make in the SFBay while only being ~25% less expensive. In real terms, I've heard that a "Sr. Engineer" is $75-85K CAD ($60K USD) where in at a typical SF startup (lower than FAANGs) you'd look at $130-140K+ USD.

Startups that get excited about low wages are rarely $B rocket ships. This makes the equity story difficult as there aren't employees who got rich from joining an early stage company. AND, the startups that did succeed also gave out less equity as it wasn't valued. Thus, a self-fulfilling prophecy.

Startups and large tech companies would get founded/move here if there's an overflow of top tier talent. There's talent but not overflowing.

Good Solutions:

* Startups should pay more.

* Toronto should attract a FAANG or Uber/Lyft/Airbnb to build a serious office here.

* Employees should hop around more.

* Import talent through Canadian visas for foreign workers (which also takes advantage of more stringent US H1-B visas)

Crazy Solution: End the TN visa forcing a FAANG to open an office in Toronto.

Originally from Toronto, lived in SF and ran a VC-backed startup from 2011-2016 and moved home to Toronto in April to found a startup doing robotaxi ops software. Plug: we're hiring.


#2 and #4 have already been done, haven't they? Amazon has a significant dev office in Toronto, and Canadian immigration rules are much more straightforward than the US ones.


A previous poster made the point of not having a major top tier tech firm with an HQ here as well. It's an interesting point I haven't thought of. I know MS is building a new big office. Google is planning a pretty large build via sidewalk labs. I need to do some research.


MS is not creating a new office. They are moving their head quarters from suburbs (Mississauga) to Downtown Toronto. They do plan to increase number of jobs from previous location though.


I'm a foreigner with ties to Toronto, and I ruled it out. Salaries vs housing costs don't add up.


I believe investors aim at ownership percentages at Series A mainly for pro-rata.

Lead Series A investors usually get pro-rata rights. Generally, the wisdom in startup investment is to double down on your winners and you typically can only do so if you have pro-rata rights. In other words, if the startup does super well, that VC will likely invest 10x more in real dollar terms to upkeep their pro-rata.

Take 2 pretend funds: CoolVC has a 20% target ownership and CheapVC has a 10% target ownership. They do their pro rata every round.

Rocketship Corp. will have the following rounds (super simplified):

Series A @ $25M post-money

Series B @ $100M post-money (15% dilution)

Series C @ $600M post-money (10% dilution)

Series D @ $3B post-money (5% dilution)

Series E @ $5B post-money (5% dilution)

Exit @ $9B

CoolVC would have exited with $1.8B + spent $100M (profit $1.7B) CheapVC would have exited with $900M + spent $50M (profit $850M)

In other words, for an additional $2.5M in the Series A, CoolVC bought an option that would ultimately make $850M more in real dollars than CheapVC.

In the VC world where 1 needle in the haystack makes or breaks your fund, it's an inexpensive option. At Series A, there should still be at least 50X potential upside.

Why do most VC funds target 15-20% ownership? Probably that's probably the most they should get to balance founder ownership through further dilutive rounds. If you look at my above example, remember that founders will probably own less than 36% of the company (they also will get diluted by employee incentive plans).


That's a fantastic explainer, thank you. In general, VC is by nature a game of ownership.


Passenger AI | Toronto | $100-160K + 0.1-1.0% equity | Full-time Onsite

I’m the founder/CEO. In 2017 I moved from startups in SF to Volkswagen in Detroit to learn about the trillion $ industry we know will transform over the next decade. I left this past April to start Passenger AI which is building remote monitoring tools for robotaxi operations. In the past 7 months, we’ve raised from top-tier VCs, built up a team of 8 (Mozilla, Lyft, Pivotal etc.) and brought on some great customers.

We moved into a new office today with a garage so we can touch/feel our product everyday.

We’re hiring engineers for DevOps, Media Streaming and ML/CV. (https://angel.co/passenger-ai/jobs)

Email me at martin@passenger.ai if you want to learn more.

No recruitment-type services please.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: