Interview with Dominic Connor

Dominic Connor is a partner in P&D. He teaches C++ in the CQF program and has been recruiting for quant positions. Based on London, UK, Dominic is well known both via his work as a headhunter and a prolific poster on the online quant communities. We'd like to thank Dominic for taking the time to sit down for an interview with QuantNet. Thanks, Dominic!

Can you give us a brief bio?

Paul Wilmott & I started P&D recruitment about three years ago. Before that I'd been a CIO for a brokerage firm in fixed income, having worked at a variety of IBs. I'm married with two kids, live in Epping Forest on the edge of London.

What is your educational background?

I'm a former student of Paul Wilmott on the CQF, and before that did Maths/Comp Sci at university, and worked on a project to apply fuzzy logic to stock picking.

What do you consider as your accomplishments up to this point?

As a teenager I worked on an industrial process to plate gold onto Teflon. Was tricky, and mildly dangerous. (hint: mixing gold potassium cyanate, nitric acid and electricity should not be tried at home). Whilst working as a consultant to the British Treasury, the accountant in charge of the British national debt calculated that on an hourly basis I was the most expensive person in the pay of the British government. This was an attempt to get the bank that employed me to cut their rates. It failed.

Any failures you'd like to tell us about?

But if you insist, early in my career I tried to apply artificial intelligence techniques to stock picking, was awful, we genuinely believed that technical analysis had some merit. My excuse was that I was young.

What other projects and/or ventures are you involved in?

I teach on the CQF, and have a venture looking at extremely low latency trading systems.

What would you want to do for a living if you weren't a quant recruiter?

I'd go back to bossing people around in some IB, or if all else fails take the job in algo trading I was accidentally offered by a bank which confused quant headhunting with being a quant. I like teaching, so would do more of that.

What advice can you give people just getting into quantitative finance field? What are the most common mistakes you see newbie making?

The most common structural error is to not making an effort to understand as much about the job as you can. A growing failure mode is students who have strong foreign accents not socially mixing with their peers outside their ethnic group. This hurts their networking a little, but in a surprising % of cases I encounter MFEs who have anything up to 4 years education in the US or Britain who I cannot understand when they speak. In some cases, my ears tell me that they're not speaking English at all. But their written and reading skills are excellent.

Another mistake is thinking that the MFE will get you a career. It will get you interviews, not even a job. You are all running up a down escalator in competition with others. You have to learn more than you are taught. Being just another MFE might get you a job, it won't make you rich.

What technical skill set, personal qualities that you look for in a prospective job seeker?

Obviously this is a math based career, and I like to see some breadth beyond the standard core that everyone is expected to have. Banks are less interested in hiring people who only know the narrow band of maths you see in quant text books. I like to see 'creativity', that's hard to define, but one approximation is the habit of trying to use ideas out of their original context. This is especially applicable to a game like ours where most people are from other backgrounds, you should be thinking of how to make this stuff work in finance. Typically you will fail, but the effort itself has value.

I like to see articulate people for several reasons. First up of course they do better at interviews, but a surprising % of the people I speak to leave me with the feeling that they don't really understand the thing they've been doing for the last few years. I could of course be mistaken, but that error hurts them more than me. But as I say elsewhere, you need to be able to explain if you want to prosper in this line of work.

How has your business changed over the years? What challenges will you face in the future?

The diversity in jobs has increased enormously. When Derman started in this business, quants were few, and most could switch track radically without effort. Now every aspect of investment banking requires at least some quant skills. Recently we talked to a large IB about quant compliance, an idea I am still getting my head around. Certainly, customer facing business development staff often now need quant skills, which hurts some older players, because they need to explain to clients and their advisors the nature of sophisticated, and thus higher margin products.

What skills (technical, social) you believe are essential to being a successful MFE?

I'm going to focus on the non-obvious aspects, rather than the obvious stochastics, PDEs, numerical methods et al.

A good technical skill, to be useful should be a lightning rod for interview questions, which means you should choose your "banner" skills as those where you can beat others.

Reviewing my posts on Quantnet, I don't push Excel VBA hard enough as a technical skill. It probably won't get you a job, but will improve your visible productivity early in your job.

In this climate, there is a stronger trend towards "needs" based hiring over "want". That means sprinkling in a few relatively low level buzzwords can make you seem like the person they have to hire to get certain work done.

Some social skills that will help you progress:
Ability to disagree with your boss in a way that does not offend him, his intelligence, or in some cases, his mother.
You're smart, else you wouldn't be doing this, and with that comes opinions which is good, but there are a pair of failure mode where quants argue needlessly and badly, and/or simply take any view of your boss with sullen silence.

Ability to explain yourself.
We like to see people who've had experience explaining things because a quant is a local expert in some area, and that means you need to communicate what you think. A good exercise is to "cross domains". For instance force yourself to explain any finance concept in terms of card games and best on horses, and when that simply won't stretch, seek out gambling games like roulette. It's news to some people that bookies hedge, which really ought to be obvious.

It's not enough to know the algebra of Kelly criteria, you should feel it, but that's no excuse for not knowing the maths.

Given current market headlines, is demand for quants increasing, decreasing, or remaining the same? Are certain quantitative skill sets becoming more valuable?

A year from now I expect more quants to be employed than today, but supply is increasing, so at some point there will be a serious gap in supply and demand. This is being delayed partly by the increase in variety of jobs

Where are the areas (product, geographic) that you are seeing most recruiting happening and who are still hiring? What will be the hottest trend?

Algorithmic trading has so far managed to avoid any major blow ups, and the technology is at a stage where the barriers to entry are relatively low. This means the large banks are not cutting this much, and some are expanding. However the quite appalling state of some major IBs infrastructure means the prop shops can compete on more equal terms. This is all good news for employment relative to the current market.

For a recent MFE graduate looking to join a trading desk, which type of desk is better, flow desk or prop desk and why? Hedge fund or IB?

You'll learn more on a prop desk, but some firms have flow trading as part of a structured process by which you get a picture of the whole business which makes you more effective. So flow is good as *part* of your training, but not so good if it's *all* you do, and prop is a steeper learning curve but you don't go so deep.

A lot depends on the IB you are comparing to a HF. I'd take DE Shaw over many IBs, but brands have more value when you are at the start of your career.

You are a big C++ advocate, how would one benchmark his level of C++ worthiness that deem acceptable to you? Is there a set of exam, test, etc?

The Brainbench tests are the least useless, but their main utility is being a cheap quick way of filtering out the truly hopeless. I regard myself as understanding some maths or programming only when I can correct the errors of others, and I think that you should not put any language on your CV unless you've fixed some bugs of your peers. This is a dirty secret of quant finance, about 60% of the work is programming and circa 50% of that is debugging. I guess a Baruch grad has 30 years of work ahead of him, so that's 10 years of fixing code. If that thought makes you physically ill, quit now.

How would one gauge his own level of success in this industry? For example, after 5 years, must one be at a certain income level with a certain title?

I'm famous for saying that you spend more time with the people you work with than who you sleep with. When people tell me about how content or unhappy they are, respect of their peers is a big factor. This is a very important and delicate issue; people who will happily tell me what they earn or important personal data, will skirt this issue.
Income correlates with respect, but again it's the rate of change and how your set of possibilities evolves that affect your morale.
I know this from my own experience, I quit my last real job because I simply could not see anything that I could do that I felt was worthwhile.

As a headhunter, it's not my job to tell people what to want, it is to find out what they want and then help them get it. Money is a factor in choosing one job over another, but for most people it is not the reason that makes them get in contact with me. Money is an objective measure of the "respect" your firm/boss has for you. One reason for the quite dreadful IT in several large firms is that IT people get random small numbers, uncorrelated with their contribution. When quant developers get dropped into IT, their morale drops faster than their pay.
One reason I'll never have to get a real job again is that nearly all banks are structurally incapable of developing people's careers. Occasionally a bank will do this mildly well, but then lose interest.

Many employers under-weight the importance of job titles as part of recognizing people's evolution

Do you have any advice for someone who may be starting their career in a second-tier financial center like Chicago?

More seriously, you need to put a bit more effort into networking, and where you have the choice try to go for more branded employers. In this market, that is not always easy, so try to pick jobs where you are learning. Chicago may be further from Wall Street, but that matters a lot less if the job is closer to the money. Sometimes there is a rational trade off to make.

What are your opinions on prop shops? I've heard people say taking a job at one of these places is a black mark on your resume as far as IBs are concerned.

Yes, some managers do see prop shops as beneath them, but many like the idea that you are close to the money, this should be something you push when you go for other jobs.
I see quite a few quant jobs right in the "high frequency" trading space, and a lot of the math involved (signal processing, etc.) are quite different than the stuff taught in typical MFE programs. Is self-study for this subjects a realistic option for these openings, or should we just leave these openings to Electrical Engineers who have been doing this for a living?

That's a pretty accurate summation, though we do have a few algo traders who are self taught in the relevant techniques. Although the strongest demand is for those with PhD level SP, good undergrad experience combined with the right maths can get you there. Typically an EE can't do more than about 25% of his degree in things relevant to SP, so self study is not impossible, but getting a job is then dependent upon your ability to project that to a prospective employer.

There is also demand for the very top end of econometrics whizzes, though the feedback I get is that employers find many mainline economists don't quite have the depth they require.

What can be done to attract more women into quantitative finance field?

Show them the money.
They also need to get the message about the work. It's inevitable that many people both males and females have an incomplete model of the opportunities and career paths. As I say above there's more types of jobs than anyone might suspect, so this is an entirely fixable situation. We advise women on this, but course if they don't look too hard at this line of work, they never find us in the first place.

The deep problem though is arts graduates in the media pumping a message that anyone who does science is an ugly geek with no friends. But we can't fix this.

What do you think are the difficulties for women to thrive in this male-dominated career and what would be your suggestions?

Although you may work for a bank with 100,000 people, investment banking leans towards lots of smallish groups, so you may end up in an all male team. Some women seem put off by this, and I have seen a couple turn down roles on that basis.
I see less networking by women, which on the whole holds them back.

There is of course sexism, and I'm not going to pretend it doesn't do damage. But I've worked in several industries and interacted with most professions, and I'm pretty confident it is less bad in investment banking than most other places, especially media, government and retail banking. I have met precisely one senior woman senior retail banker, yet like in nursing there's plenty at the lower levels.
But the money in quant work is better, so my cynical position is that if you're going to be discriminated against anyway, you might as well get well paid for it. IBs make more millionaires from ethnic minorities than any other line of work, so I don't see any major barrier to extending that to women.
We're looking at doing an "Investment Banking for Girls" event with one of the major banks, for women doing the right subjects. Aside from the requirement to think of a better name, it will use all our resources to get enough in one place simply because the vast majority of proto-quants are men.

What have been your experiences placing female candidates? Do they face more obstacles or being short-changed as a female?

It's easier for us to get an interview for a woman, and on average they are more likely to get a given job. Being a headhunter, it follows that I have better data on the transitions between jobs than how women progress within those jobs. Hiring managers from a wide range of business have privately told us they'd actively like to see more women. That to me is a clear signal that if I find more good female quants, my business will grow faster.
Joining the dots of the snapshots I have of female quant careers, I see no evidence that FQs move up the food chain more slowly or earn less. Although my data is sparse enough that there may well be a bias, it tells me that it is not very large, because I'd spot that.

There is a shadow of a signal that FQs get lower bonuses, certainly they convey that idea to me, and this implies some degree of lower appreciation of their work.

Kids are an issue of course, and if someone leaves this line of work, I can't figure them into my model. I have to choose my words carefully here, but it can be made to work, if you want it to. Mrs. DCFC is a partner in a very large international law firm, earning money comparable to an IB quant, and so we have a nanny, and various other fixes available to those who earn relatively well. The wide variety of jobs means that a working mother has a respectable set of options. I accept there is an impact (DCFC 2.0 and 2.1 cost more than a new house in lost earnings), but when I see mothers in lower paid jobs outside banking they don't have so many choices and those options are often much less attractive. An FQ might go and work for a consultancy and gain better hours at the price of "only" earning 4 times what the average person does.

If you were asked to design a curriculum for MS program in Financial Engineering which core courses would you pick?

The core of most MFEs is the same, and I have no issue with that. The structural change I'd look at is more deep-specialist options. As the market tightens, it will be less useful to know the same general stuff as everyone else. That is a long term trend as well, as the size of this profession grows then you cannot hope to cover everything in adequate detail.
Thus I'd identify some niche like energy derivatives, risk management or government debt and build excellence all the way from trading through to cutting edge modeling and hedging. I'd split roughly 2/3 core competencies in time series, pricing, numerical methods, etc, and 1/3 making people stand out as first rate in that area. Done properly you can attract the top names to guest lecture and work with students, and of course banks wanting that particular skill will come shopping.

If you were asked to run an MFE program, what would you do to place your students for internships and jobs?

As a headhunter, I see the alumni as the most powerful resource for getting my students into firms. I would make sure that we kept contact with them, with things like continuing education, lunches, meetings in bars. Create a yearbook with a summary of the program and the CV of everyone who's looking for a job. Include a list of the subjects covered by students and the depth they did them to. Mail it to all alumni. None of this costs much.

I'd also have a short series of lectures specifically aimed at the skills relevant for getting a job. That's conditional probability, brain teasers, how to explain what you're good at, and CV design.
Also I'd encourage the alumni to work on the recruitment system. Headhunters are the first gatekeepers in the process, and try their best to find out what clients want. If they think Baruch students sell well, you will find yourself being put forward to more roles. It's also important to know that in this line of work you are interviewing people much earlier than elsewhere. This can actually be just a few weeks, and very probably you will interview a newbie within your first year. Thus there are Baruch alumni getting resumes from headhunters all the time. They just need to ask "why am I not getting many Baruch people ?". The HH will get the message, and if enough do that it will swing the consensus your way. I'm not saying Baruch alumni should give preferential treatment to own school in hiring decisions which is wrong on so many levels, just poke the information providers to see the picture better.
(Just don't tell the other HHs I told you to do this).

What do you know now that you wish you'd known 10 years ago?

That I could do this for a living, is certainly the first item. I'm lucky that I enjoy my job, enough that my 7 year old son has difficulty understanding that it is work.

Tell us something about yourself that we don't already know.

I've started writing a novel about Rogue Trading, based upon the realities we've seen, with various tales I get to hear as a headhunter.

The working title is "Fat Tails", maybe "Trade/Off" what do you think ?

What does the future hold for Dominic Connor?

I ought to do a careers talk at Baruch, and physically meet the people who read my posts on Quantnet.

See also
I think what many people are forgetting is that foreigners use the school system to get their foot in the door in America. By taking their graduate courses here they are allowed to find a job within a certain period of time. This causes a disproportional amount of foreigners in graduate programs. Also, you forget that the population differences in India and China versus the United States can cause a seemingly disproportionate amount of math oriented people when you get into the work force in America.

Granted, the American primary and seconadry public school systems are terrible. You could probably point to the parents and their seemingly uninterest in their children (you have parents blaming schools that their high school senior kid cannot read), teachers are trounced for kids not learning but learning takes place at home and needs to be instilled. The majority of Americans are lazy. They don't read. They don't think for themselves. etc. etc. Even for gifted students they rarely allow children to skip a grade for fear of being a social outcast. I have even seen a few cases where a kid is very smart and talented but is bored with how slow the schooling is going. So of course he turns to being disruptive and being "off-task" and then people proclaim he has ADD and is put on medication!

(Before anyone asks... yes I am a white American male and am also upset with the way this country is running itself into the ground.)
Sounds about right. After finishing high school with math-physics concentration, I went into acting. Had no idea what to do with math skills. It didn't look like a career path with many opportunities.
Incidentally, the Associated Press just published a study showing that girls are just good as boys when it comes to math.
Parents and teachers persist in thinking boys are simply better at math, said Janet Hyde, the University of Wisconsin-Madison researcher who led the study. And girls who grow up believing it wind up avoiding harder math classes.
"It keeps girls and women out of a lot of careers, particularly high-prestige, lucrative careers in science and technology," Hyde said.
News from The Associated Press


Quant Headhunter
I don't believe the AP thing about girls and maths.

Two reasons. In cultures where there is no particular push towards boys, they still do worse.

But the big thing is sampling AP and other news reports.

Examine the next 10 reports on "sex A is better than sex B at ...."

I will bet you real money that every single one of them says Women are better programmers, drivers, pilots, carpenters, sheep herders etc.

One day soon I expect to see a survey that says most gay men say that women make better lovers.

Maybe women are better at maths if taught the right way. Could be.
Maybe women are better on a larger set of distinct measurable skills than men, certainly from an evolutionary perspective that makes some sense (sorry to use the "E" word on an American forum, but I'm a Brit and under recent legislation all English science grads are required to sneer at least once per month on social sites.)

How credible is it that 99% of all testable skills are found to be higher in women ?
I would have been a UFC fighter

I grew up in India and off course encouraged to study math and science. I did it pretty good, got into IIT (for people who know what it is) and have received couple of awards for contribution in engineering, etc. But when I first saw the UFC fight I felt if I was not encouraged/coerced to choose my current career this is what I would have been doing. Don't misunderstand my current love for MFE. I feel I am/will be definitely happier juggling finance equations in C++ than I would have been playing with punching bags.
... indeed, this is not only in finance ...
There is indeed an oversupply of the talent you refer to - typically those with excellent math+programming skills. But bear in mind there is still an acute shortage of people who can apply these skills to generate profit, which is essentially the whole point of financial engineering.