Apr 2024 – Jun 2024 · 2 mos · Gurugram, Haryana, India · On-site

• Worked on to implement a generic end-to-end C++ infrastructure (completed and tested on NSE), which included market data decoder, bookbuilding, strategy API and Exchange Order Sender. Identified a number of bugs in existing infrastructure and replaced some strategies with the new infrastructure implemented by me. Significantly improved tick to trade latency.

Jan 2024 – Apr 2024 · 3 mos · London, England, United Kingdom · On-site

• Market Access Team - Fixed some bugs in their Solarflare TcpDirect based transport library used for communicating market data to strategy. Also worked on decoders and book building of some exchanges - OPRA, Euronext and CME. Redesigned some legacy classes which used function pointers for callbacks, to use CRTP instead.

Jul 2023 – Jan 2024 · 6 mos

• Execution Algorithms (Manager) - Analyse and implement new EA strategies for optimal slippage in different markets. Eg. - Percentage of Volume (POV), SmartPOV, Bracket Order, Smart Order Routing.
• London Stock Exchange Dev - Developed Market Data and Order Entry applications for LSE on the native protocol.

Apr 2023 – Jul 2023 · 3 mos

Sep 2022 – Apr 2023 · 7 mos

• Execution Algorithms (Team Lead) - Managed the deployment of EA as a generic service integrated with GUI which can now be used by all teams to send orders or square off via TWAP / VWAP algo. Also, extended EA to multiple exchanges.
• Core ORS - Implemented restart recovery protocol which corrects ORS and strategy state upon restart without any manual intervention. Before this, upon ORS restart, all strategies connecting to it had to be restarted by trading teams.
• Simulation - Extended ML Delay to multiple exchanges (OSE, ICE, Eurex) and streamlined the generation of Dynamic Delays (recorded delays) for all exchanges we are currently trading on.

Jul 2021 – Sep 2022 · 1 yr 2 mos

• Simulation - Implemented an end-to-end framework (from feature generation to prediction) for ML Based Delay calculation in Simulation from scratch. Added support for new crypto exchanges and a framework to easily register user provided functions inside simulation.
• Managed 2 interns to complete the following projects in a span of 11 weeks - ag-scheduler and Central Server (handles app scheduling and configuration dependencies), Processed Data Compression (new algorithm to store historical TBT data, 30% size reduction), ML Delay Prediction Pipeline for simulation.

Jul 2020 – Jul 2021 · 1 yr

• Execution Algorithms - Analyse and implement new EA strategies for optimal slippage in different markets. In addition to handling existing teams, successfully on-boarded 5 new teams on 4 new exchanges within 3 months.
• Alphaqueue - A distributed data streaming and storage library used to communicate trading information across 100s of machines in real-time. Complete redesign to make robust and more maintainable.
• Tomac - Backend for GUI application used by traders. Improved restart-recovery mechanism by handling dropcopy trades and added support for square-offs.

May 2019 – Jul 2019 · 2 mos · Bengaluru, Karnataka, India

• Synonym Sense Mapping - Perform Sense to Sense mapping across Synonyms using Google’s Dictionary Data. Changes are currently live at Google Search Page.

May 2018 – Jul 2018 · 2 mos · Noida Area, India

• Paraphrase Generation - Built a Deep Learning based model which when given an input sentence, generates various other sentences with similar meaning.

May 2017 – Jun 2017 · 1 mo · Bengaluru Area, India

• Trinity - Built a recommendation service using Graph Traversal Algorithms (Implemented using Neo4j)
• Monroe - A convenient WebApp for data feeding by Fashion experts.