Were you told that you couldn’t learn algorithmic trading?
Were you told that you wouldn’t be able to make it out there?
Even after such remarks from naysayers, you’ve strived, worked hard and reached your challenging goals.
This is also reminiscent of a story that resonates with all the aspiring Algo Traders out there. The story of a Commerce graduate who became a Data Scientist and learned algo trading.
An NLP Researcher, an Azure Certified Data Scientist, an AWS Certified Cloud Practitioner and now an EPAT alumnus, Karthic has a passion for turning data into actionable insights, meaningful stories and implementing what he learns.
This is his journey into the world of algorithmic and quantitative trading.
Hi Karthic, tell us about yourself!
I am Karthic Krishnan, I’m from Bangalore and I’m working as a Data Scientist at IBM. I’m from the Commerce background, hold an MBA in finance and also a PG in Business Analytics.
Even with this varied education, I was able to transition, my career actually from business development, and marketing into data science, and it has been quite a lot of learning.
I’m also an NLP Researcher, an Azure Certified Data Scientist, and an AWS Certified Cloud Practitioner.
I love to play chess and often trek with my friends to rejuvenate from the daily work life.
How did you transition from BCom to the Financial and Technical side?
Being a BCom graduate, I was passionate about research. Back then, algo trading was not that familiar in India, but I was thinking of building an automated trading system, creating strategies and such.
At that time, life had other plans. I did my MBA in finance. Life demanded, so I took up a sales and marketing profile - a quite interesting functional area. I would interact with clients, know about the product.
My first job was actually into financial services and that's how I got into trading. Wanting to get into technical research, I got an opportunity in Chennai, India where I was learning to code, exploring candle charts and different patterns.
Being from a non-Computer Science background was slowing down my progress. To achieve my aim and to transition my career to set up a trading desk, I needed:
- a lot of coding knowledge
- to dig deeper, dive into the research
This transition was a little painful because of initial negative comments from a well known close circle of people, but I got good support from my family to try something new.
I was in my mid-30s, with 2 kids, and it was a tremendous pressure to leave my job and enroll myself into a full-time post-graduation program in Business Analytics in Praxis Business School, one of the top Business Schools in Analytics Business Analytics Program.
It was a rigorous program of nine months to one year. I put in all my efforts like spending 14 to 16 hours a day completely learning to code, exploring different platforms, different programming languages like Python, R Studio, Sass, visualization tools like Qlik Sense, Tableau, Database System, SQL and whatnot. That's how I acquired strong coding knowledge.
I’m also passionate about statistics, so I thoroughly brushed up on advanced statistics, mathematics, data science concepts, algorithms, and complex algorithms.
I was able to TOP in three subjects in the overall batch! I never realized that I became proficient with coding during the program, because I was the only one at the time in my batch without a computer science background, or any certificates.
This completely changed the way I perceived challenges. After that, I was able to see those challenges be it personal or professional, and win systematically. That's how actually, the transition happened.
Was it challenging to embark on a career journey to today becoming a Senior Data Scientist at IBM?
With a good score in my PGDM, I got into campus placements, landing a job at Hinduja Global Solutions as a Junior Data Scientist. Some years later, I got promoted, and after completing a big milestone project, I got an opportunity at IBM. Thus, I’m working as a Sr. Data Scientist at IBM for more than 2 years now.
It's a very challenging and demanding role requiring expert skill sets. One has to handle complex problems, work on complex algorithms and scaling, etc. This has moulded me. I've mentored junior data scientists and hired capable and productive candidates. Over these 2 years, I was able to realize that using my coding skills I could solve this Market Prediction challenge.
When I started applying my techniques to solve business challenges, I wanted to get more hands-on experience before I take any programme or course. I was searching for something that could help me pursue my path. I came across a couple of programs out of which, QuantInsti’s EPAT was very impressive. And that’s how I entered this world of Algorithmic Trading.
Entering EPAT, I was quite curious. So, I started the learning process quite early, as I knew to code. I took the data freely available in the market. I was thinking about:
- which market to focus on,
- fetching expensive nifty data, and all of that.
But, EPAT put my anxiety to rest.
The curriculum is fantastic. EPAT extensively covers every subject from length to breadth and end to end whether it's risk management, portfolio management, or the ratios to evaluate the model, different strategies, etc. EPAT gave me good lateral and vertical knowledge.
How have you been applying the skills and knowledge from EPAT?
I visited the Binance site and got 3-4 days of freely available data, and built a model by writing 8000 lines of code. I utilised my Data Science skills to download data, prepare the data, and then develop machine learning models. I tested several models before that.
I also worked on an NLP based project that helped me a lot to work on this input parameter post which I’d work on the strategy. Because it's time-series data, any machine learning model should compensate for the market’s dynamic changes.
I’d need to evaluate the model stability. So, I created my own parameter to evaluate the model’s stability - if the model is doing well, and if it will do well in the future. If it doesn't do well in the future, I can retrain the model, and change the parameters.
On top of these strategies, I tried several strategies, which I learned from EPAT and I also modified them. I had worked on all this a long time back when I was in the Technical Research team.
It all came back to me when I started working on the model. The strategy helped me a lot, it's a different thought process altogether. And that's how I built the model only to realize that my models gave profitable trades.
I couldn’t possibly allocate a server or infrastructure for non-stop functional cryptos - definitely not while only my job, considering I’m working on projects. So, I developed the model, trained it and hired freely available AWS Cloud Services. I deployed it on the cloud without any additional cost and it runs consistently.
For me, the journey of 10 to 12 years from the financial services, as a technical analyst, to data scientists over here now, it’s quite interesting. I never looked at EPAT as an opportunity, in terms of a great career setting, but I saw it as an opportunity of turning a passion into something in which I can tangibly get the results. And it has supported me a lot.
Being a Data Scientist, how would you describe your experience learning Algorithmic Trading with EPAT?
EPAT is a fantastic programme, comprehensive and quite good. The following are my perception and experiences of EPAT.
Market Prediction is a challenge that has multiple factors and has millions of brains working towards it. This requirement nudged me towards EPAT.
With EPAT, I became studious over six months. I never thought this much weightage in the curriculum would be given to Data science concepts. I was familiar with some aspects, but the other aspects have a strong value addition for me. Plus, I wanted to strengthen my knowledge.
Other courses vs EPAT - EPAT is more intuitive as it consists of strategy building, it's a combination of statistics data science, algorithms and other technical evaluations of the model, and also the governance. Other courses just provide strategies that don’t fit everyone as it keeps changing.
Comfort for me - As I was thorough with coding, statistics and data science concepts, it was comfortable for me to go along with the programme, although it was very robust. However, there was a LOT to learn.
Applied what I learnt - Being from a commerce background, I was able to recollect those concepts - Omega, Sharpe ratio, etc. It was surprising to see variations coming up in the model.
Helped me explore the domain more! It's up to each individual how deep they go into each subject in the programme. As an experienced Data Scientist, I’d dig deep to learn how to make some hyperparameter modifications, and changes. All and all to build my knowledge.
Alumni benefits - EPATians are still connected via the Alumni group where rich and useful information is shared.
Faculty - Faculty is one of the key features of EPAT. Because they are already experts and have a strong market background, deep knowledge and they have seen different aspects of - be it institutional trading, HFT, etc. They are very much patient enough to clarify each and everyone's questions. I observed that I could not see such attention to candidates in all the various courses and programmes that I’ve been in. For each faculty, the candidate’s query is important.
Placement - The placement team and services are wonderful. The services will be extremely beneficial for everyone from freshers to 3-4 years’ experience individuals and more. I saw many offers coming up in data engineering, data science, data scientist, and also the quant developer and quant analyst roles.
Time management - Work was hectic sometimes due to multiple projects, so I was able to spend time on the weekends via the weekend lectures. I was able to manage a work-study-life balance all the while supporting my family.
Thank you, QuantInsti for providing this rigorous yet wonderful programme - EPAT.
I liked All of the features of EPAT. It's quite impressive!
What do you have to say to those people who wish to learn Algorithmic Trading?
For the people who are not from a coding background or don't have any other data science background and wish to get into this domain, I'm the best example.
Tough times never last, but tough people will.
Never give up - The journey never stops.
Never have self-doubt.
I had zero technical knowledge, coding knowledge, and I was not from an engineering background. I did not have those right skills; I faced a lot of humiliation to start with. But today, I'm working in IBM as a Senior Data Scientist mentoring a team.
13 years back, I started my carrier as a Sales Executive. I have stood in the stalls, on the kiosks, handing out pamphlets in temples, and supermarkets, etc. I’ve crossed all those days. I’m not underestimating those profiles but saying that I've been through all of these roles. But, I kept trying and never gave up.
Never expect that someone will come and give you a flower and say, “You will do it”. The first person to believe in you is only you! That's the key thing.
So, anyone who has a fire to transition their carrier, actually, to algo trading and data science,
- keep focusing
- learn the fundamentals
- have a healthy thought process
- be very confident
- be patient
- be curious
- getting the required skill sets
- have a plan
- have a structure
Definitely, success will be all yours. Definitely, you can succeed.
Thank you for sharing your journey with us, Karthic. Your journey was filled with a lot of challenges, but you whisked them away with your sheer determination, will and desire to learn. You’ve proved that one can indeed accomplish their aims if one truly wishes to. We wish you the best in your future endeavours.
If you too desire to equip yourself with lifelong skills which will always help you in upgrading your trading strategies. With topics such as Statistics & Econometrics, Financial Computing & Technology, Machine Learning, this algo trading course ensures that you are proficient in every skill required to excel in the field of trading. Enroll in EPAT now!
Disclaimer: In order to assist individuals who are considering pursuing a career in algorithmic and quantitative trading, this success story has been collated based on the personal experiences of a student or alumni from QuantInsti’s EPAT programme. Success stories are for illustrative purposes only and are not meant to be used for investment purposes. The results achieved post completion of the EPAT programme may not be uniform for all individuals.
