Monthly Archives: March 2012

Here Are The Women of Y Combinator And They Are Awesome | TechCrunch


I would normally rather have a root canal instead of write about the issue of women in technology. I just find most essays on this really tedious and obvious. (Sorry Alexia.)

But I do want to point one thing out. When I went to my first Y Combinator Demo Day three years ago, there was one woman. At this week’s Demo Day, there were four companies with one or all female founders among the 66 startups in the class.

I’m going to keep this post simple. No complaining. Less navel gazing. Just more role models. So here are the women of Y Combinator and they are awesome.

(Ladies, if you’re interested in joining the next class, the deadline just passed. But there are two classes a year, so the next one will come up soon.)

Elli Sharef, HireArt

Growing up in Colombia, Sharef was lucky to have a strong female role model right by her side. Her mother had a Ph.D. in economics

“She’s a strong figure with opinions and she was an intellectual,” Sharef said. “I never thought about being a man or woman. She just told me to be ambitious, to do my thing and try and build something good for the world.”

Sharef’s company is attacking the HR and recruiting space. She’s a co-founder of HireArt, which is trying to ease that first step of sifting through an impossible number of resumes.

HireArt has job candidates actually perform a series of tasks or do video interviews. For example, if an interview candidate says they are an expert in Excel, they can demonstrate their skills on HireArt by creating an Excel model using a dataset.

“I saw how hard it was to hire the right person. Everyone knows that the right person can 10X your team,” she said. “At the same time, it’s equally bad when you don’t hire the right person. It can be really terrible.”

HireArt’s site is growing 40 percent week over week and currently has 238 open positions. The company earns revenue every time a candidate is successfully placed, the way a good recruiter might earn a fee or a salary percentage if they find a good hire.

To get into Y Combinator, Sharef came together with a few friends from her university days at Yale: Dain Lewis and Nicholas Sedlet.

There’s a question of how easy it will be to scale HireArt’s model given the idiosyncrasies of hiring and finding a good cultural fit between employees and employers.

Sharef says that over time, the company will collect more and more data from employers about interview questions or tests that are strong predictors of success.

“We really try to work with data to understand which questions work the best. You can think about it like designing the SATs for different jobs,” she said, pointing out that one of her co-founders has experience working with huge data sets as a commodities trader and quant.

4 is better than 1, but out of 66, that’s still a long way to go. Ladies, get to founding!


DRACO: Death to the Virus

DRACO: Death to the Virus

by  —  November 17, 2011


In a paper published 27 July [1], researchers from MIT reported successful tests in mice with a new drug that holds the promise of being a cure to all viruses. The drug, DRACO (Double-stranded RNA Activated Caspase Oligomerizer), works as a “broad-spectrum” antiviral, killing virus-hijacked cells by targeting double-stranded RNA produced in the viral replication process. DRACO proved successful against all 15 viruses tested “including rhinoviruses that cause the common cold, H1N1 influenza, a stomach virus, a polio virus, dengue fever and several other types of hemorrhagic fever.” [2]

We may expect results from cell trials against AIDS within the next 12 months.

DRACO is but one broad-spectrum therapeutic being developed as part of a project called PANACEA (Pharmacological Augmentation of Nonspecific Anti-pathogen Cellular Enzymes and Activities) headed by Dr. Todd Rider, senior staff scientist in MIT Lincoln Laboratory’s Chemical, Biological, and Nanoscale Technologies Group.

I met with Dr. Rider in the food court of the MIT co-op bookstore early on a weekday. He had already finished tending to his mice and, after we chatted, he rose to declare that he off to do “real work”… writing grant proposals to keep his research alive.


Could you give us a broad overview of the Panacea project?

Sure. We’ve come up with a broad-spectrum antiviral that we call DRACO, Double-stranded RNA Activated Caspase Oligomerizer (I love acronyms), and it’s basically designed to detect any long double stranded RNA, so we’ve created chimeric proteins where one end will detect the chimeric RNA — the double-stranded RNA — and then the other end will trigger apoptosis, or cell suicide. So the net effect is that these DRACO molecules can go inside all the cells in your body, or at this moment, inside all the cells in a mouse, and if they don’t find anything, then they don’t do anything. But if they find a viral infection, if they find a viral double-stranded RNA, then that will activate the back ends to trigger cell suicide, and that will kill the infected cell. That terminates the infection.

So there wouldn’t be a difference between DNA Viruses and RNA Viruses?

It works with both. We’ve tested it on both. All known viruses make double-stranded RNA, and that’s true from the literature and also true from our experiments. So here (indicating illustration) the viruses we tested included a couple DNA viruses, and it worked quite nicely against those. Others in the literature are also known to make quite a bit of double-stranded RNA. Other DNA viruses, like pox viruses and herpes viruses, also make double-stranded RNA.

Has it been tested on each family of virus?

It’s been tested on these families of viruses so far (indicating paper). There are a gazillion viruses, so we’re working our way through them as quickly as we can. It’s been tested on several very different families so far.

My understanding is that viruses usually kill the cell anyway, but retroviruses usually do not. I don’t know how viruses cluster. Are there any odds at all that there would be a retrovirus that clusters too tightly in a certain organ where it [triggered cell death by DRACO] would cause a lesion?

Virtually all viruses will kill the host cell on the way out. Of the hand-full that don’t, your own immune system will try to kill those infected cells. So we’re really not killing any more cells with our appraoch than we already have been. It’s just that we’re killing them at an early enough stage before they infect and ultimately kill more cells. So if anything this limits the amount of cell death.

So that’s not really a legitimate fear.

It shouldn’t be.

How far along are you and how far away are you from human trials?

Unfortunately quite a long way. We’ve done a number of tests in mice. We need to do more testing in mice. Of course, MIT is not a pharmaceutical company. There’s only so far we can take it at MIT. We’re hoping to license it to some pharmaceutical company, and they would carry to larger-scale animal trials. Usually the FDA wants to see a lot of mouse trials, which we’ve done already; and then a lot of trials in, say, rabbits or guinea pigs, and then trials in monkeys before they approve human trials. So, if a licensee takes this, if we have funding for it, it still might take a decade or so before it really is available for humans.

So how’s the funding working now?

We have funding from NIH [National Institutes of Health].

And can you take it up to monkey here [at MIT]?

We may be able to take it into further animal models here, but mice are the easiest thing to use. We have a lot of mice. We’re also limited by funding. We only haved NIH funding at the moment, and we only have enough funding for about 1 person, and we have 4 people total, counting me, working at the moment, so we’ve split the funding four different ways…

Has anybody reached out to you?

Nope. Not so far.

When I first read about this I thought this was an amazing story, that this would be front-page news in a couple of hours. Weeks later, I was thinking this must not have been a true story. That’s when I looked it up again and saw that it was indeed on the MIT site. What’s the relative lack of interest. There haved been articles, but I feel this is definitely front-page material.

Well thank you. On the funding front, I think there’s a ton of funding for very basic research — not applied research, trying to cure something, but basic research — Let’s go study this virus, see how this virus works in a little more detail. There’s a ton of NIH funding for that. On the applied front, if you are ready for human trials — so you’re 10 years more advanced than we are now — then there are government agencies and companies that will take it and take it to that final step. But in that long gap in between there’s very very little funding out there. So we’ve been struggling for all of 11 years now just working to get funding, and at the moment we’re just barely limping along.

This is a subset of PANACEA, right? Can you describe PANACEA?

PANACEA is a family of broad-spectrum anti-pathogen treatments. We’ve tested some others, we’ve tried to get funding for others. This [DRACO] is the one that is furthest along.

What are some of the others that look promising?

We have a number of others. [DRACO] is a broad-spectrum antiviral. We have other broad-spectrum antivirals. We also have other PANACEA treatments that we’ve adapted to go after other things. Like for bacteria. And of course there are antibiotics, but for bacteria that are resistant to existing antibiotics, such as tuberculosis, malaria… so we can adapt this to pathogens other than viruses. We’ve done some initial experiments, we just can’t get funding for that so far.

Do you foresee any potential wild-cards in the human trials?

It’s always difficult to tell what will happen. I hope that there won’t be. We’re always concerned that there will be some toxicity or other unforeseen problems. We’ve been very pleased every step of the way in the cell testing. We’ve tested in a number of different human cell types representing many different organs; human lung cells, human liver cells, all kinds of different human cells, as well as a variety of animal cells. We haven’t seen any toxicity or any other strange effects in any of those cell types. In the mice we were again very concerned about toxicity, and we haven’t seen any toxicity in the mice. We inject the mice with very high doses of the stuff daily for a number of days, and they seem fine. We let them move for a while, eventually we dissected them, looked at the tissues. All the tissues were fine, there’s no organ damage or anything. It’s always possible something unexpected could come up further down the road in monkies or in humans. We certainly hope not. But I think there is enough flexibility in the concept that even if there were a problem, there are ways to redesign the constructs that we have to overcome any potential problems.

That might also speak to the production cost. Is it fairly low production cost if, say, it was to be mass-produced in the future?

These are produced in bacteria, and at the moment I really don’t know what the ultimate production cost would be. We produce on a very small scale, barely enough for our mice. Of course cells eat a lot less DRACO than mice do. So if we’re producing for cells, that’s a very small quantity, but just a few flasks of bacteria will produce enough to last us for a while. But once you scale this up to a large-scale production large-scale animal trials or human trials, hopefully the cost would go down. I don’t know exactly what the cost would be.

Do you envision the final end-plan to be people with DRACO in their medicine cabinet, or more like penicillin today?

If it’s safe I’d like to see it used as much as possible for as many different things as possible. I would guess that if it were approved for human use by the FDA, initially they would be conservative enough that they would only want to see it used in very dire cases, just in case there are interesting side-effects or something, and it’s only to people with ebola or HIV that’s become resistant to other drugs who would get this. If this proved to be safe in those cases, then I would hope that they’d approve it for wider use against more common pathogens, perhaps all the way down to the common cold. And if it really is safe, then maybe you’d just pop a DRACO pill any time you felt a cold coming on.

How long does it stay in the system? It’s obviously not a vaccine –

Right. In cells it lasts at least for a couple of weeks, possibly longer. In the mice it lasts for at least 2 days. We have a lot of data in the paper showing it will persist in mice for at least 48 hours at fairly high doses in the tissues. This is really about trying to optimize that. There are a lot of tricks we can use to try to make it last longer if necessary. And if this stuff is truly completely safe, then you can give it prophylactically. You could even concievably give someone the gene for the DRACO so that their cells would just permanently produce the DRACO, and they would naturally be resistant to almost everything.

Oh, wow. That’s an amazing idea.


I feel like this is something that should be fast-tracked. We have all this planning in regards to epidemics. There is all kinds of scare that we’re ripe for an epidemic.

Perhaps we will be [approached with funding offers] in the future, but so far we haven’t been. We’ve really struggled along for the past 11 years, barely getting enough funding to stay alive.

So this has been on the table, at least as an idea, for 11 years?

Right. We just got good data from the mouse trials and published that, but 11 years ago we started engineering the DRACOs. Genetic engineering was a bit more primative in those days, so it took us a while to actually produce these things. Then it took us a while to produce and test them in cells. We ultimately tested against 15 different viruses in cells. As I said, we were kind of limping along for funding for much of that time, so we could only work on it when we had funding to work on it. For some fraction of our time, we had funding to work on it. Eventually, we were able to test against the 15 different viruses in cells in 11 different cell types. And then we had funding to do some mouse trials, got data, and then we got published.

If you get a cold this winter… are you going to be tempted?

I’m not tempted by colds. I’ve had very bad stomach viruses and I’ve been tempted to give myself the stuff to see what would happen.

You don’t think you’ll do that, though?

It wouldn’t be enough anyway. We only produce enough for mice, and for a human you require a much larger dose than for a 20 gram mouse.



Marked as: ScienceTechnology  —  14 comments   (RSS)


14 Comments so far

  1. CogitoErgoBibo November 20, 2011 12:06 pm

    Thank you very much for following up on this landmark work. I think we would all be served well if we can get this viral (no pun intended) on the net and I intend to look into some method to get non-corporate funding to help them. I’ve never tried, but this is that important in my opinion.

    I am self-taught in the bio sciences starting on Immunology last year. It has been demonstrated widely that protein function is modular. That is parts of proteins (domains) having a particular function can be “inserted” into other proteins to add function and, further, completely new proteins can be generated. This discovery has led to many innovations and great potential for medicine.

    Past this, there are astronomical numbers of untried amino acid sequences. Even if we consider a small proteins of 100 amino acids in lenght, there are more sequence combinations than there are atoms in the universe. Nature only selects for what is immediately necessary, now, our protein engineers with a process called Directed Evolution, we can select for sequences that can save lives and improve life in general for everyone.

    I started writing on what I learn and I have even found a research path of my own. I use a blog as a repository, if its appropriate I’ll comment again with the link. (I didn’t want this message to be seen as spam)

  2. doug November 20, 2011 6:14 pm

    @CogitoErgoBibo —

    Thank you. Feel free to post your links, I’m curious to see what you’re working on.

    Thanks –


  3. CogitoErgoBibo November 20, 2011 7:06 pm

    Thanks Doug.

    A quick leadup… While watching Immunology webcasts from the NIH, I came across Dr. Carl June speaking about Chimeric Antigen Receptors and, specifically, about his clinical trial during which 2 chemorefractory patients with CML achieved a complete response in 30 days (cancer free).

    At the time, I started thinking about how we could get around the major limitation of his landmark work, but given that I am self-taught and all of it was new to me it took a few weeks… Anyway, I developed an idea and sent it to one or two people in the field just in case it might spark an idea in them… they encouraged me to continue. So, a few months and a few hundred pages of research later…

    Thank you again for your attention to DRACO and your interest in my efforts.

  4. CogitoErgoBibo November 21, 2011 6:41 pm

    I’ve been posting the link to this post everywhere I can think of… even sent it to my Congressman. I am putting together a plan as this is something I have to at least try to remedy.

  5. CogitoErgoBibo March 18, 2012 5:28 pm

    Updates on two fronts… I did get a letter back from Rush Holt, my congressman, and he said he would “advocate for higher levels of funding for this important priority”, but of course it could have just applied to “medical research” which was the subject of the previous sentence.

    On my research, an article I co-authored with a prof from the University of New Haven was published here and I was interviewed by their Personalized Medicine TV site here:

  6. jhaukur March 29, 2012 11:26 am

    Here is an idea.

    Use Kickstarter for funding!

    Game development companies have been raking in a couple of million dollars through kickstarter,

    but a cure for all viruses?

    I’d donate, so would others.

  7. Nunsuch0 March 29, 2012 1:59 pm

    There’s now a new petition to White House to fund the further development of this potentially momentous drug:

  8. Ycros March 29, 2012 2:06 pm

    I agree with the Kickstarter comment, I would happily throw in some money if something like this was setup.

  9. doug March 29, 2012 3:07 pm

    It would certainly be interesting to see how much grass-roots funding could be generated. Unfortunately, Kickstarter has policies which exclude projects like Draco research. They reject anything involving “drugs”, and while it could be argued that they are referring to illegal narcotics, they also exclude “Health, medical, and safety-related products”. Let me contact Dr. Rider for an update on the current state-of-affairs and I’ll tell him about the interest being generated in a grass-roots fund. I’d love to help set something up…

  10. Ycros March 29, 2012 3:12 pm

    What about Indiegogo or others?

  11. doug March 29, 2012 4:51 pm

    I’ll take a look at it. Maybe we’ll be able to set up something more directly still….

  12. Nunsuch0 March 29, 2012 6:15 pm

    I’ll be watching this page for any updates on direct funding… Though, I don’t know what are the MIT policies on donations – is it possible to fund a certain project directly, or would it have to go into communal pot. Anyone knows?

  13. CogitoErgoBibo March 29, 2012 6:28 pm

    Thank you Doug for providing the knowledge and the forum for getting the eyes of smart people on what I think is a critical piece of research.

    I have spoken to a friend in IT Venture Capital about starting up a NFP to channel funding for Dr. Rider’s work. I also asked about investigating patent acquisition if that’s possible. The pipedream, of course, would be to add revenue into the funding stream, then look for other research opportunities proactively rather than reviewing grant proposals as is customary.

    I’d be happy to help in any efforts to get this work funded.


  14. CogitoErgoBibo March 29, 2012 6:29 pm


    I’ll call Lincoln Lab tomorrow and post what I learn here…


Leave a Comment

If you would like to make a comment, please fill out the form below.

You must be logged in to post a comment.

This is fascinating stuff. Can it really be so simple that all viruses have a common weakness, and one that is so fundamental to their functionality that they can’t evolve their way past the weakness?

Global Warming Close to Becoming Irreversible: Scientific American


“irreversibly?” Really? The evidence does seem to indicate that the climate is getting warmer and less stable. It does seem to indicate that we’re behind it. I understand the concept of tipping points, and Greenland (to pick one aspect) is a really big chunk of ice to try to put back if it melts. But in the words of Dr. Sheldon Cooper, “irreversible” is an absolute, and using it in this way is sloppy and alarmist.

Not having a perfect memory sucks sometimes

I remember seeing 12 Monkeys back in 1995 and being very frustrated with some aspect of the time travel portrayed. For years I’ve been unable to remember exactly what it was that ruined the film for me, but I knew there was something. I put off watching it again because I knew it was so frustrating.

I finally broke down and watched it again, and damn it, I still can’t remember what I was pissed off about. 

Not that there aren’t obvious issues (spoilers ahead):

If someone is gunned down in an airport and the person with them is screaming out that that someone getting onto the plane is planning to release a deadly virus, would the plane still take off?


Correction: having typed the below, I’m now pretty confident that I know what pissed me off, and this is it:

If you are in the past (prior to 1996) and trying to convince people to take your warnings about the future seriously, how hard would it be to make predictions that would make it clear you know what you’re talking about? You might not be able to convince them immediately, but if you named a few presidents and a few wars, it seems clear that it wouldn’t be too hard. The film makes it clear that at least one time traveler ended up 600 years in the past, at least one ended up in the early twentieth century, and one ended up in 1990.

The film was based on the premise that you can’t really change the past: that everything that will happen to your future self in the past already has happened, and is unchangeable. So for example, when Bruce Willis was a child he saw his adult self get killed. There was never a time (sorry to overburden that word) when his child-self didn’t see his adult-self killed because it hadn’t happened “yet.”

Unfortunately, that’s just silly on the face of it. It’s hard to imagine a logically-consistent timeline where people from the future have free access to the past, free will to act/communicate in the past, but no ability to affect it. If someone in 1920 wanted to call out Hitler, would it be so hard to name enough “future” events to be taken seriously, and also call out Hitler? Is there any self-consistent (i.e. unchanging) timeline where someone could be in that position, say what was going to happen, and it not change events? Could a world that had access to a detailed chronology including the stock market crash, Presidents Hoover and Roosevelt, The Jazz Singer, and Shirley Temple (those from memory) not take seriously the subsequent predictions of the rise of Hitler and Mussolini, the Japanese attacks on China and Pearl Harbor? And if the world did, how could those events play out as they did?

If that were the case, the historical timeline in 12 Monkeys would have to be far different than our own, having been affected by numerous time travelers. But if that’s the case, how could time travel be unknown in 1990 and 1996? Combine that with the doctrine of “what will happen has happened” and you get a fixed timeline where many people have gone back in time, but none has managed to do anything interesting while back there. Either no time travelers have made a serious effort to influence anything, or they have somehow been prevented. The former is clearly not true, even based just on Bruce Willis’s character, and there is no evidence of the second idea, which was so well explored by Fritz Leiber in a series of short stories.

If there’s a way for a timeline to be non-self-modifying, time travelling, information-sharing, change-motivated but change preventative, I don’t know what it is.

Recruiting: 8 Qualities Your Best Employees Should Have

8 Qualities of Remarkable Employees

Forget good to great. Here’s what makes a great employee remarkable.

8 qualities of remarkable employees

shutterstock images

Great employees are reliable, dependable, proactive, diligent, great leaders and great followers… they possess a wide range of easily-defined—but hard to find—qualities.

A few hit the next level. Some employees are remarkable, possessing qualities that may not appear on performance appraisals but nonetheless make a major impact on performance.

Here are eight qualities of remarkable employees:

1. They ignore job descriptions. The smaller the company, the more important it is that employees can think on their feet, adapt quickly to shifting priorities, and do whatever it takes, regardless of role or position, to get things done.

When a key customer’s project is in jeopardy, remarkable employees know without being told there’s a problem and jump in without being asked—even if it’s not their job.

2. They’re eccentric… The best employees are often a little different: quirky, sometimes irreverent, even delighted to be unusual. They seem slightly odd, but in a really good way. Unusual personalities shake things up, make work more fun, and transform a plain-vanilla group into a team with flair and flavor.

People who aren’t afraid to be different naturally stretch boundaries and challenge the status quo, and they often come up with the best ideas.

3. But they know when to dial it back. An unusual personality is a lot of fun… until it isn’t. When a major challenge pops up or a situation gets stressful, the best employees stop expressing their individuality and fit seamlessly into the team.

Remarkable employees know when to play and when to be serious; when to be irreverent and when to conform; and when to challenge and when to back off. It’s a tough balance to strike, but a rare few can walk that fine line with ease.

4. They publicly praise… Praise from a boss feels good. Praise from a peer feels awesome, especially when you look up to that person.

Remarkable employees recognize the contributions of others, especially in group settings where the impact of their words is even greater.

5. And they privately complain. We all want employees to bring issues forward, but some problems are better handled in private. Great employees often get more latitude to bring up controversial subjects in a group setting because their performance allows greater freedom.

Remarkable employees come to you before or after a meeting to discuss a sensitive issue, knowing that bringing it up in a group setting could set off a firestorm.

6. They speak when others won’t. Some employees are hesitant to speak up in meetings. Some are even hesitant to speak up privately.

An employee once asked me a question about potential layoffs. After the meeting I said to him, “Why did you ask about that? You already know what’s going on.” He said, “I do, but a lot of other people don’t, and they’re afraid to ask. I thought it would help if they heard the answer from you.”

Remarkable employees have an innate feel for the issues and concerns of those around them, and step up to ask questions or raise important issues when others hesitate.

7. They like to prove others wrong. Self-motivation often springs from a desire to show that doubters are wrong. The kid without a college degree or the woman who was told she didn’t have leadership potential often possess a burning desire to prove other people wrong.

Education, intelligence, talent, and skill are important, but drive is critical. Remarkable employees are driven by something deeper and more personal than just the desire to do a good job.

8. They’re always fiddling. Some people are rarely satisfied (I mean that in a good way) and are constantly tinkering with something: Reworking a timeline, adjusting a process, tweaking a workflow.

Great employees follow processes. Remarkable employees find ways to make those processes even better, not only because they are expected to… but because they just can’t help it.

Jeff Haden learned much of what he knows about business and technology as he worked his way up in the manufacturing industry. Everything else he picks up from ghostwriting books for some of the smartest leaders he knows in business. @jeff_haden

I don’t always exhibit these characteristics, but I do more than I used to.