Principal Engineer

Our trading platform is the engine the fund runs on. We're looking for the engineer who will own it — the architecture, the code, and the team that builds it. You will set the technical direction and make the hard architectural calls, and you will stay deep in the code, building the most critical and latency-sensitive parts yourself. You will lead a strong engineering team, and you will be accountable for everything the platform does in live markets: that it's fast from tick to trade, that it's correct, that it stays up while markets are open, and that it behaves safely with real capital.

This is a builder's leadership role. You own the outcomes across the whole platform — including its connectivity, colocation, resilience, security, and regulatory posture — and you will shape the infrastructure and the exchange, broker, and vendor relationships beneath it. What this is not is an IT-administration job: as we grow, dedicated infrastructure and operations support will carry the hands-on run-work so you stay focused on building and leading.

Responsibilities:

• The architecture and technical direction of the platform end to end — market data, signal, execution, and risk — and the calls that shape how it's built.
• The hardest code. You design and build the critical, latency-sensitive components yourself, and set the engineering standard the team builds to.
• The tick-to-trade path: finding and winning back latency across ingestion, decision, and order placement, and keeping it predictable under load.
• Risk and safety in the system — the pre-trade and at-trade controls that keep the platform from doing damage with real money. These are the controls that have to hold even when everything else is failing.
• Own the testing and staged-rollout discipline that makes changes safe to ship.
• The team — hiring, mentoring senior engineers, owning delivery, and raising the bar.
• Reliability in live markets: uptime through the session, fault tolerance that fails safe, monitoring of both the systems and what they produce (fills, PnL, slippage), incident response during market hours, and the hardening that follows.
• The platform's connection to the outside world — exchange and market-data feeds, the OMS, brokers — and accountability for the leased lines, colocation, physical infrastructure, redundancy, and failover beneath it, including getting new venues (NSE, BSE, MCX) certified and connected and coordinating the exchanges, brokers, and vendors behind them.
• Resilience and continuity — backup connectivity, disaster recovery, and business- continuity planning, tested with real drills.
• Security and governance — setting the security and governance policies, building access control, secrets, and auditability in, and keeping the platform ready for SEBI and exchange scrutiny.
• Scale — growing the platform with data, accounts, strategies, new markets, and AUM, and partnering with research and operations to turn what they need into systems that hold up.

What Success Looks Like in the First Year:

• You know the platform cold, and the team ships through you without you being the bottleneck.
• Tick-to-trade latency and uptime are measured, understood, and trending the right way.
• The risk controls are something you'd trust with the firm's capital — because you've tested them.
• A new venue or market has gone live cleanly, or the path to it is genuinely de-risked.
• The platform's security and DR posture would survive a SEBI or exchange inspection without a scramble.

Requirements:

• A senior engineer who leads from the front — you want to own both the architecture and a team, and you're better at each because you do both.
• Someone who can personally solve the hardest problems in a real-time system, not route them to someone else.
• An owner by temperament — of decisions, of uptime during market hours when something breaks, and of the consequences.
• Calm under the particular pressure of live markets, where mistakes cost money, and the clock doesn't stop.
• A clear communicator who can work across research, operations, and compliance, and hold their own with exchanges, brokers, and vendors.
• 10+ years building and operating production backend systems, with real time spent leading engineers or owning a platform's technical direction.
• Expert Python and deep command of asyncio and concurrency — event loops, backpressure, race conditions — on an async end-to-end platform (FastAPI/Starlette, heavy asyncio).
• A real track record with real-time, event-driven systems: live market-data and order- event streams over WebSockets, coordinated through a pub/sub backbone.
• Experience integrating with external trading infrastructure — exchange and market- data feeds, an OMS, and brokers — over socket and HTTP APIs, with the discipline order placement and state tracking demand.
• Strong judgement across a mixed data layer: an operational store such as MongoDB, Redis (with Sentinel for high availability), and a time-series or columnar store such as ClickHouse.
• Proven ability to design always-on services for reliability and observability, and to own uptime in production.
• Comfort owning a containerised, multi-service setup end to end — Docker, CI/CD, and automated deployment.
• Enough grounding in the Indian market structure and regulatory context — exchange connectivity and colocation, OMSs such as XTS, Greeksoft, or ODIN, venues such as NSE, BSE, and MCX, and SEBI's IT, cyber, DR, and BCP expectations — to own these outcomes and work credibly with exchanges, brokers, and vendors.
• A record of making sound architectural decisions and leading a team to deliver against them.

Nice to Have:

• Experience building high-frequency or ultra-low-latency trading systems, where the hot path is measured in microseconds.
• Hands-on with kernel-bypass networking (Solarflare/Onload, DPDK), FPGA, or other hardware acceleration for market data and order entry.
• Experience with tick-by-tick (TBT) feeds and the colocation setups that make microsecond latency achievable.
• Low-level performance engineering: lock-free data structures, cache-aware and zero- allocation hot paths, and C/C++/Rust where Python can't go.

Skills

Asyncio, Concurrency, FastAPI, Greeksoft, ODIN, OMS, Python, Starlette, Trading, WebSockets, XTS