TL;DR
Trading firms hiring graduates from IIT, ISI, BITS and similar institutes look for four things: fast, accurate probability and mathematics, clean programming, evidence of self-driven depth in one area, and the ability to explain reasoning clearly under pressure.
The institute opens the door. What decides the offer is how a candidate performs in timed tests and how convincingly they discuss their own work.
Key points
- Speed with accuracy beats breadth of coursework.
- One deep project beats five shallow ones.
- Honesty about what you did and did not do is scored positively.
- Communication is part of the technical assessment, not separate from it.
Signals that carry weight
| Signal | Why firms value it |
|---|---|
| Probability fluency | Central to research and trading decisions |
| Competitive programming results | Evidence of problem-solving speed and correctness |
| Olympiad or contest performance in mathematics | Strong indicator for research roles |
| A real project with a write-up | Shows independent work and honesty about results |
| Systems depth | Valuable for engineering roles: operating systems, memory, networks |
| Research internships | Shows sustained work on an open-ended problem |
| Clear explanation | Teams need people who can be challenged and respond well |
Signals that are overrated
- Long lists of online courses with no applied work.
- High grades alone without evidence of depth in something specific.
- Finance certifications for research and engineering roles.
- Trading app screenshots presented as investing experience.
- Backtests claiming very high returns, which experienced interviewers read as inexperience.
How to present a project
- State the question you were trying to answer.
- Describe the data and its problems.
- Explain your method and why you chose it.
- Give the result, including what failed.
- Say what you would do next with more time.
This structure works for research, systems and hardware projects alike, and it is exactly how firms expect their own people to report work.
What firms assess by role
| Role | Primary signals |
|---|---|
| Quant research | Probability, statistics, research judgement, Python |
| Quant trading | Mental maths, decision-making, risk instinct, temperament |
| Software engineering | Data structures, systems knowledge, code quality |
| Low-latency engineering | Memory and performance awareness, curiosity about hardware |
| FPGA | Digital design coursework or projects, verification thinking |
See the HFT skills matrix for depth by role.
What gets candidates rejected
- Memorised puzzle answers without understanding.
- Inability to explain code they wrote.
- Claiming credit for group work.
- Rigid thinking when an interviewer challenges an assumption.
- Careless errors in timed tests, which cost more than slow answers.
How to stand out without a perfect CV
- Do one project thoroughly and write it up publicly.
- Compete in mathematics or programming contests and share your approach.
- Learn one thing deeply, such as order book mechanics or Linux performance.
- Apply through open processes and be specific about why the firm interests you.
- Ask thoughtful questions in interviews about the work itself.
After the offer
Early-career quants and engineers usually gain independent work within a year or two. The candidates who progress fastest keep the same habits that got them hired: precise reasoning, honesty about results and curiosity about how markets and systems really behave.
Frequently asked questions
Do trading firms only shortlist certain branches?
Shortlists often favour mathematics, statistics, computer science and related branches, but firms also consider strong candidates from other branches with relevant evidence.
How important is CGPA for HFT jobs?
It can affect shortlisting, but test and interview performance usually matters more.
Does competitive programming guarantee an offer?
No, but strong results are a well-regarded signal, particularly for engineering roles.
What do interviewers think of trading side projects?
They are welcome when presented honestly, with realistic results and clear discussion of costs, data quality and validation.