jasontan

Sift Science - San Francisco, CA. Full-time.

Sift Science (http://siftscience.com) uses large-scale machine learning to fight online fraud. It's a problem that cost U.S. merchants > $10B last year, and 70% of it is organized crime. Attacks have rapidly evolved in breadth and depth, but current rule-based systems don't scale. We're looking for engineers of all flavors -- distributed systems, web development, data visualization, and of course, machine learning. We're a tight-knit team that likes board games, yummy food, and solving challenging technical problems. Check out https://siftscience.com/jobs We're also looking for account managers, integration engineers, and someone to lead our B2B marketing efforts.

Feel free to email me personally - jason at siftscience dot com

Thanks for the kind words!

Sift Science - San Francisco, CA. Full-time.

Sift Science (http://siftscience.com) uses large-scale machine learning to fight online fraud. It's a problem that cost U.S. merchants > $10B last year, and 70% of it is organized crime. Attacks have rapidly evolved in breadth and depth, but current rule-based systems don't scale.

We're looking for engineers of all flavors -- distributed systems, web development, data visualization, and of course, machine learning. We're a tight-knit team that likes board games, yummy food, and solving challenging technical problems. Check out https://siftscience.com/jobs

We're also looking for account managers, integration engineers, and someone to lead our B2B marketing efforts.

Feel free to email me personally - jason at siftscience dot com

Hey Eran! We actually help quite a few e-commerce sites today. To be clear, 99% of our scores are updated and ready within a second of receiving the transaction. The fact that the response to a transaction does not include this score is a design flaw, and you're not the first to bring it up. We're working on a fully revamped API that should address this, and the revamped API will be launched in the next 2-3 months.

I do remember your emails from back in the day -- will followup offline.

10minutemail.com and yopmail.com too! :-)

Yep! We try to localize the time.

Hey Aaron, I'm the OP, and CEO of Sift Science. Sorry that your experience has been subpar. I'll follow up offline.

I think there's some confusion -- we actually do have a global machine learning model, which is what customers will start with if we haven't received any labels (training examples to learn from). But, your mileage may vary with the global model. We think it's a starting point, but that fraud differs from site to site in subtle ways, and to be truly effective you need to have your own model. This is why training examples are so critical. Once we receive enough training examples, we're able to build a model specific to your site.

That said, I'll be the first to agree with you -- we haven't done a great job setting the right expectations and messaging, especially on the point above (that your mileage may vary with the global model). We've been working hard to improve this, and there are now several reminders in the console, and an introductory tour, that emphasize the importance of a thorough integration with labels to achieve best results. With machine learning especially, bad data in = bad results out. Also, our first UI/UX designer started a month ago, and improving this experience is a top priority.

With regards to making fraud facts more functional and actionable -- we actually do provide reasons we think a user is suspicious in our web console and API (see https://siftscience.com/docs/getting-scores). But, there isn't an "aggregate learnings" page like what's presented in the blog post -- that's on the 3-month roadmap.

Again, sorry for the subpar experience, and I'll follow up with you separately. We have a very technical product -- we're working hard to abstract the complexity, while making sure we set customers up for success. We've also been overwhelmed with customer demand, and have been scaling the team to keep up (7 FT employees two months ago, 15 now). That said, no excuses. Just know that we're not happy to hear about experiences like yours, and are working hard to make it right.

Thanks, Jason

(EDIT) Aaron and I just had a good phone call -- we're going to make things right.

How is this the right thing? I don't imagine that your former coworkers are having a good day today with this kind of publicity. They had nothing to do with this matter, and did nothing to deserve the bad attention.

I understand that it's important to get the word out, but for me, I'd put the people I care about above anything else. All I'm saying is that this could have been handled more professionally.

While this is unfortunate and it sounds like you deserve better (I'd like to hear the other side of the story first), I would discourage against taking this public. This is an issue between you and the founders, and no one else. "Warning" your former Miso coworkers should happen privately -- this post brings them the wrong kind of attention, but they did nothing to deserve it.

Future employers who read this post will think twice about hiring you -- frankly, this is rather unprofessional. Yes, the Miso founders might be in the wrong, but you can still take the high road.

Just signed up. Looking forward to it!

most importantly, Sahil went to Singapore American School.

I spent fourteen years in Seattle until I moved to SF this June. I sorely miss the natural beauty. The air is cleaner, the nature is encompassing, and people are smart and down-to-earth. The rain is manageable -- just don't use an umbrella, otherwise people will know you're new to the city :-)

I've heard that you and your founders were all serial entrepreneurs going into Heroku. What did you learn from your previous companies?

You should do a little Jig :-)

Yep, this is true. A friend of mine is producing one of their channels (deal has been signed). Exciting!

if you can't get an introduction, continue working on your product and gain traction, if you have a million users you definitely won't be ignored.

Definitely appropriate, for this community even more so. Steve was an inspiration, an icon, a legend.

He built the iCloud so he could have a place to rest in peace. Thanks, Steve

awesome, thank you. now if only we could do split panes...

congrats oliver

ironically, we ran into PG & JL downtown right after he did the interview. while we were lugging cinema displays...

Path-on-ura

Time for a new startup: Path-on. Social sharing of pictures of coupons. Mobile x realtime x commerce. Already valued at $2B. Who's with me?

It was really confusing to understand what I was looking at until I read the blog post. I'd recommend screenshots and/or a "how-it-works" diagram explaining what you do. I'm hesitant to "sign up" without knowing what I'm getting into. I like the concept though.

my early feedback -- this seems like a natural candidate for using Facebook authentication. you could post to the user's Wall "I just read X" and there'd be a natural viral factor.

driving traffic... most startups have that problem. HN, reddit, digg, email relevant bloggers, tell your friends (that's without spending anything).

making money -- seems like ads would be the way to go. if this took off, i can also see "sponsored content" mixed into the recommendations, ala "promoted tweets".

good luck!

Someone from http://www.teamgantt.com commented on my RMS, it looked pretty legit...

no, you absolutely make sense. other comments also point out this problem. previously, Find My Buzz was more like an "instant feedback" product, instead of a wait-for-your-emailed-report -- the animation sequence was displayed while they waited. we ditched the "instant" part, but thought the animation was too cute to let go... alas, we must do what's best for the user!

It appears that we can do a better job of messaging that this product is intended for local businesses with a physical address. Yes, the zipcode is for disambiguation and ensuring that we only include mentions relevant to that business in that location.

We have a separate product that better serves multi-unit chains and "brands" (http://www.buzzlabs.com), but that is not as self-serve as Find My Buzz.

Thanks for the feedback on the screenshots, definitely a common theme here and we're fixing that now.

we originally tried a "real-time" model and only asked for an email once the report was ready -- but this turned out to still be too slow and caused the user to wait too long. asking for the email upfront is our way of setting expectations on when they can receive the report.