About

I've been a software engineer for most of my career. Here's a quick summary of the jobs I've been fortunate enough to have over the last 10 years:
  • I was the second non-founding engineer at LinkedIn. The story of how I lucked into that role can be found here. While at LinkedIn, I worked on the first versions of products like LinkedIn Groups and LinkedIn Jobs and also implemented some of the core graph algorithms. 
  • After two years at LinkedIn, I spent 3+ years at Google, where I spent most of my tenure working on real-time payment fraud detection. I wrote a large portion of the software infrastructure for that project and also designed and implemented some algorithmically complex fraud signals. 
  • My most recent job involved spending 4 years at Factual, a location data platform. While there, I built many components from the ground up including Continue reading "About"

Best Of

Data Analyses

The Value of Data

Seed Stage Logistics

Pitching and Fundraising

Fundraising

Book/talk notes

Ways I Can Help

(This page is intended for current/prospective portfolio companies.)  I often tell people, "let me know if I can help with anything," but what does that really mean? In general, I think founders should ask investors for anything they can think of. That said, here's my attempt at an exhaustive list of things I'm good at and things I'm not so good at. Overall philosophy...
  • I help whenever possible and try to get out of the way when I can't. My goal is to be a resource, not an anchor.
  • I try to reach out proactively every month or so, but please feel free to email me/get in touch with me whenever you'd like.
  • I'm honest about what I do and do not know. If I don't know the answer to your question, I will tell you that instead of just making stuff up.
I can provide feedback on...

Archive of posts

Best Of

Data Analyses

The Value of Data

A series about building valuable datasets and creating business models around those datasets. Parts: 1, 2, 3.

Seed Stage Logistics

Pitching and Fundraising

Fundraising

Book/talk notes

About

I've been a software engineer for most of my career. Here's a quick summary of the jobs I've been fortunate enough to have over the last 10 years:

  • I was the second non-founding engineer at LinkedIn. The story of how I lucked into that role can be found here. While at LinkedIn, I worked on the first versions of products like LinkedIn Groups and LinkedIn Jobs and also implemented some of the core graph algorithms. 
  • After two years at LinkedIn, I spent 3+ years at Google, where I spent most of my tenure working on real-time payment fraud detection. I wrote a large portion of the software infrastructure for that project and also designed and implemented some algorithmically complex fraud signals. 
  • My most recent job involved spending 4 years at Factual, a location data platform. While there, I built many components from the ground up including an in-memory distributed search engine; a logging and analytics service; and the bulk of the data cleaning, deduping, and normalization pipeline.

I've been working with three great partners on Susa Ventures since late 2012.

Ways I Can Help

(This page is intended for current/prospective portfolio companies.) 

I often tell people people "let me know if I can help with anything," but what does that really mean? In general, I think founders should ask investors for anything they can think of. That said, here's my attempt at an exhaustive list of things I'm good at and things I'm not so good at.

Overall philosophy...

  • I help whenever possible and try to get out of the way when I can't. My goal is to be a resource, not an anchor.
  • I try to reach out proactively every month or so, but please feel free to email me/get in touch with me whenever you'd like.
  • I'm honest about what I do and do not know. If I don't know the answer to your question, I will tell you that instead of just making stuff up.

I can provide feedback on...

  • your product (UX/UI, new features, website copy, new mobile app builds, etc).
  • product direction and features ideas.
  • pitch decks.
  • sales decks.
  • general company strategy.
  • any type of written work: blog posts, one-pagers, etc.
  • etc.

I can provide high-level software design assistance with...

  • API design and API feedback.
  • Written product specs, API specs, and API docs.
  • System design and system design feedback.
  • Working through use cases for pushing a product into production.
  • Advice on infrastructure design decisions (which NoSQL datastore to use, etc). I'm not super experienced here, but sometimes I know enough to be useful.

I have practical software domain expertise in...

  • strategy and implementation details around data cleaning, deduping/entity resolution, data enrichment, etc.
  • designing data processing pipelines (advice about high-level design rather than specific infrastructure decisions).
  • fraud detection (and related problems).
  • feature design for ML problems. I did a lot of this while working on fraud detection at Google and on deduping at Factual.
  • I don't know ML, but I'm great at algorithms/heuristics/etc. I love thinking about hard/intractable problems and am pretty good at solving them (I used to place well in national and international programming competitions).

Ways I can help with recruiting...

  • forwarding potential candidates whom I meet if they seem like a good fit (if you tell me what you're looking for).
  • screening/interview process design (I did this a lot at my last job).
  • recruiting strategies and messaging (I did this a lot at my last job).

Intros I can provide..

  • my network is so-so (I've lived outside of the bay area for most of the last 5 years), but I'll do my best to connect you with whomever you are looking for.

Miscellanea

Archive of posts

Why You Can’t Hire

There isn’t a shortage of developers and designers. There’s a surplus of founders. The cost of starting a company has collapsed. It’s now just (minimal) salaries. For entrepreneurs, desks are free, hosting is free, marketing is online, and company setup is cheap. Raising the first $25K for product development is easy – join an incubator. Raising the next $100K is easy – investors are following the incubators with automatic notes. Building a product and launching a product are easy – develop on Open Source Stacks, host on Amazon, launch on Facebook, Android or iOS, get your early traction.* Getting real traction is hard. Raising millions of dollars is hard. Building a sustainable, long-term company is hard. Yammer can hire. Square can hire. Twitter can hire. These companies have achieved product / market fit. Your pre-traction company has not, and so it has a hard time hiring. If the costs
Continue reading "Why You Can’t Hire"

Why You Can’t Hire

There isn’t a shortage of developers and designers. There’s a surplus of founders.

The cost of starting a company has collapsed. It’s now just (minimal) salaries. For entrepreneurs, desks are free, hosting is free, marketing is online, and company setup is cheap.

Raising the first $25K for product development is easy – join an incubator. Raising the next $100K is easy – investors are following the incubators with automatic notes. Building a product and launching a product are easy – develop on Open Source Stacks, host on Amazon, launch on Facebook, Android or iOS, get your early traction.*

Getting real traction is hard. Raising millions of dollars is hard. Building a sustainable, long-term company is hard.

Yammer can hire. Square can hire. Twitter can hire. These companies have achieved product / market fit. Your pre-traction company has not, and so it has a hard time hiring.

If the costs of founding a pre-traction company have gone down, then returns to pre-traction founders must go down.

Throw out the old cap tables. A founder doesn’t get 30% and an early engineer shouldn’t get 0.25%. Those are old numbers from when you had to raise VC capital before you could build a product. Before everyone could and did start a company.

Post-traction companies can use the old numbers – you can’t. Your first two engineers? They’re just late founders. Treat them as such. Expect as much.

Your next five designers and developers? Your cap table probably can’t even afford them until you have traction, and the cash that follows it.

Close the equity gap, and hiring will get a lot easier.


* Of course nothing is ever “easy” – but it’s a lot easier than it used to be.

** This is just my opinion, not that of my employer. But you can see what they’re doing to help at AngelList Talent. Coincidentally, the lead hacker on that project put up this related must-read post on his own blog yesterday.


Archimedes Labs opens for business

<img class="alignright wp-image-509" title="Lever.png" src="http://i2.wp.com/www.archimedeslabs.com/wp-content/uploads/2011/12/Lever.png?w=150" alt="" px" data-recalc-dims="1" /> Palo Alto, CA, 12 December 2011 Today is an exciting day at Archimedes Ventures. We have some announcements to make. Firstly, going forward we will be known as Archimedes Labs LLC. This is to make clear that we are an incubator, not a venture capital company. Secondly, we can now announce that in August this year I established a managing entity – Escondido Labs LLC. It has been the sole manager of Archimedes since that time. Thirdly, we can reveal that in the future investors will be added at Archimedes Labs as partners in the entity. Finally we are announcing the following members (aside from myself) at Escondido Management LLC: Kambiz Hooshmand – Former senior executive at Stratacom which Cisco acquired for over $4 billion in 1996; Cisco SVP; and CEO at Applied Micro (AMCC), Kambiz is now CEO and a General Partner at Escondido Management and runs the management of Archimedes Labs. Patrick Gannon, a founder at Lending Club, Patrick is Chief Investment Officer at Escondido Management. Kevin Doerr, VP and Chief of Staff (Products) at Yahoo. Matt Kaufman, VP Product and Engineering at Oodle. Bess Ho – Mobile Architect, Lecturer & Book Author. See the team page for more details. In the first 11 months of 2011 we have incubated just.me, entered into an acceleration agreement as a co-founder at Blurtt and invested in Quixey, Incident Technologies and Broomstick Productions. In 2012 we intend to do more incubations and accelerations. Happy Holidays to all….

Towards a Literate Nation

“Give me a lever long enough, and a place to stand, and I will move the earth”

– Archimedes

To reduce unemployment, we need to lever up.

Not like Wall Street did, through debt, but like Silicon Valley does, with tools.

Leverage magnifies your actions and increases your productivity. You can get leverage through:

  • labor (people work for you)
  • capital (money works for you)
  • tools (machines work for you)

We’re trying to help labor. We don’t have much capital. We must give tools to the people.

All of the great modern tools for productivity – the printing press, the factory, the movie studio – require capital and coordination to use. The computer is the first tool since maybe the stone axe, that an individual can use to gain massive leverage, without permission from anyone else.

The modern computer can help in every endeavor – even if you don’t use a

Continue reading "Towards a Literate Nation"

Towards a Literate Nation

“Give me a lever long enough, and a place to stand, and I will move the earth”

– Archimedes

To reduce unemployment, we need to lever up. Not like Wall Street did, through debt, but like Silicon Valley does, with tools. Leverage magnifies your actions and increases your productivity. You can get leverage through:
  • labor (people work for you)
  • capital (money works for you)
  • tools (machines work for you)
We’re trying to help labor. We don’t have much capital. We must give tools to the people. All of the great modern tools for productivity – the printing press, the factory, the movie studio – require capital and coordination to use. The computer is the first tool since maybe the stone axe, that an individual can use to gain massive leverage, without permission from anyone else. The modern computer can help in every endeavor – even if you don’t use a
Continue reading "Towards a Literate Nation"

Towards a Literate Nation

“Give me a lever long enough, and a place to stand, and I will move the earth”

- Archimedes

To reduce unemployment, we need to lever up.

Not like Wall Street did, through debt, but like Silicon Valley does, with tools.

Leverage magnifies your actions and increases your productivity. You can get leverage through:

  • labor (people work for you)
  • capital (money works for you)
  • tools (machines work for you)

We’re trying to help labor. We don’t have much capital. We must give tools to the people.

All of the great modern tools for productivity – the printing press, the factory, the movie studio – require capital and coordination to use. The computer is the first tool since maybe the stone axe, that an individual can use to gain massive leverage, without permission from anyone else.

The modern computer can help in every endeavor – even if you don’t use a computer at work, you’ll soon carry a $50 smartphone in your pocket. You’re banking online, learning online, communicating online. The computer, and now the smartphone, make everyone more productive.

This is why Silicon Valley doesn’t have enough people, when the rest of the nation doesn’t have enough jobs.

Now computers are simpler. More ubiquitous. Cheaper. More accessible.

And we can teach people how to use computers, through computers, using tools like ShowMe, CodeAcademy, Bloc, and so on.

Let’s create the world’s first, completely Digitally Literate Society. Not a nation of programmers (maybe someday) but a nation of people who are comfortable with the most powerful tool ever invented by mankind.

Let’s do it quickly – train them in three months from start to finish. And cheaply – for free.

We can create the program at Stanford, MIT, Berkeley, Harvard, etc. We can teach basic proficiency – use online applications like Google Apps, Search, DropBox, Email, Online Banking, Travel, Ordering supplies, etc. Some basic creation – create your own web-site. Research questions. Solve problems. Learn to learn.

Google, Amazon, and Apple will loan us the tablets. Companies will pay us to have people learn to use their apps. An app marketplace where companies pay society to educate society.

Apple, Square, DropBox, Twitter have created beautiful software interfaces. iPad and Android have made the hardware accessible. Today, it is not just cheap to build a company. It is cheap to re-build a person.

Let’s create the world’s first digitally literate society. The world’s most desirable workforce. If that won’t generate employment, nothing will.


What I’m Excited About for 2012 (and Beyond)

In keeping with compulsory end-of-year traditions, I wanted to share what I’m most excited about as an early-stage technology investor heading into 2012 and beyond. If 2011 was all about VCs losing their heads in the consumer Internet craze, I think 2012 will be about the re-emergence of enterprise investing which has been a black sheep for many venture capitalists since around 2008. (Perhaps nothing makes it clearer that enterprise investing is on its way back than this Pulitzer-worthy piece of journalism from TechCrunch: http://tcrn.ch/tx0S6b).

So, with no further ado, here are several themes that I’ll be spending my time exploring in the coming weeks, months and years:

Democratization of Big Data
I view the democratization of Big Data as two things: 1) giving any organization – from enterprise to startup – a cost-effective way to harness the power of the data it generates and 2) to empower any user within the organization with tools that make it easier for him or her to glean insights from that data.

The first statement reflects the point that startups today generate petabytes of both structured and unstructured data but do not have the financial and/or technical wherewithal to deploy multi-million dollar Netezza, Vertica, Greenplum, Teradata, etc. appliances and data warehouses. Hadoop offers some benefits (e.g. runs on commodity hardware) but it’s still costly and complex to deploy and manage. Amazon Elastic MapReduce offers access to Hadoop clusters as-a-Serivce, but outside of that, smaller companies today don’t have affordable, out-of-the-box or on-demand solutions to process, analyze and then gain insights from the data they generate. In 2012 this will start to change.

The second point reflects that for an organization to make sense of their data they often are forced to make at least two or three specialized, expensive hires: a Java Engineer to set up the Hadoop infrastructure, a Data Engineer to write the data processing algorithms and then a Data Scientist to generate insights from the processed data. This is too much overhead. That’s why I’m excited about products like Platfora that unlock Hadoop and enable Business Analysts with little or no technical expertise to derive insight from massive sets of data.

Cloud (Security of)
“Cloud” has been a buzzword for more than five years now, but really only in the last two have organizations started to realize the immense potential from this paradigmatic shift in how compute resources are structured and utilized. What’s slowed adoption, however, has been concern over security, compliance and governance of data in the cloud. Incumbent network security and security software vendors have not innovated adequately to meet the demands of this new architecture, but in the last year or two there seems Continue reading “What I’m Excited About for 2012 (and Beyond)”

Holding your hand to the fire

Several months ago, I wrote a post about the four stages of learning.  Implicit in the transition out of stage 1, the "enthusiastic beginner", is the fact that to really learn something, you need to try to do it, and inevitably, do it wrong.  

Given the wealth of fantastic content out there in the blogosphere, especially when it comes to starting a company, raising capital, building a product, etc., it can be tempting to try to read blogs and hope that they'll help you navigate potential road mines. But if you are brand new to any one of these domains, most of what you read won't really "catch" in your brain. Instead, I find the best time to learn something is after I just messed up on that thing.

So if you're starting something for the first time, and find yourself making rookie mistakes like building too many features into the first version of your product, or not having your "elevator pitch" ready when you pitch a VC, don't worry.  To really learn, you need to hold your hand to the fire.

As Rob May of Backupify once tweeted, "Everything I learned about doing a startup came from doing a startup... not reading blogs about it."