Senior Network Engineer - Trading Infrastructure

Keka Technologies Private Limited; Bengaluru; On-site; INR 3,500,000 - 7,000,000; Full time

Job summary

Keka Technologies Private Limited in Bengaluru seeks a Senior Network Engineer to own the design, build, and operation of our trading network and server estate, delivering ultra-low latency across colocation sites and exchange connectivity. This hands-on role combines networking, systems engineering, and trading infrastructure.

You will tune Linux, automate provisioning across colo sites, and own the observability stack for latency telemetry and incidents.

Qualifications

• 7+ years in network/systems engineering including 3+ years in HFT or ultra-low-latency environments.
• Deep hands-on expertise with low-latency switching, multicast, BGP/OSPF, and exchange connectivity.
• Strong Linux internals knowledge - kernel tuning, performance profiling, and troubleshooting under production load.
• Experience optimizing server hardware/firmware for latency.
• Configuration management and deployment at scale: Ansible, Puppet (or Salt/Chef), PXE/kickstart provisioning, golden image and firmware lifecycle management, Git-based change workflows, and CI-driven rollout.
• Monitoring and observability with Prometheus and Grafana; familiarity with other tools like Zabbix, Nagios, Telegraf/InfluxDB, ELK/Loki.
• Experience operating production systems where downtime has direct financial impact.

Responsibilities

• Architect, deploy, and tune low-latency switching fabric (Arista 7130/7150, Exablaze, Cisco Nexus) and exchange cross-connects.
• Manage exchange and market-data connectivity: multicast feeds, A/B feed arbitration, order entry sessions, and colo cage buildouts.
• Build and tune trading servers end to end - BIOS/UEFI settings, C-states, P-states, turbo, hyperthreading, NUMA topology, and PCIe lane placement.
• Own Linux kernel tuning: IRQ affinity, CPU isolation and shielding, hugepages, scheduler and NIC driver tuning, and kernel bypass (Solarflare Onload, Exanic, DPDK).
• Automate server provisioning, configuration, and deployment across colo sites - no snowflake boxes, everything reproducible from code.
• Own the monitoring and observability stack for network, servers, and latency telemetry, including alerting and on-call escalation.
• Deploy and maintain precision timing infrastructure: PTP/White Rabbit, GPS grandmasters, hardware timestamping.
• Build latency measurement pipelines - tap aggregation, packet capture, and correlation of wire-to-wire and tick-to-trade metrics.
• Lead capacity planning, vendor and carrier management, and new venue onboarding.
• Mentor and grow a small team of network and systems engineers; set standards and run change control for production trading systems.

Skills

Low-latency networking

Linux performance tuning

SRE / on-call

Automation / IaC

Vendor & carrier management

Mentoring

BGP/OSPF

Networking hardware optimization

Latency measurement

Tools

Ansible

Puppet

Salt

PXE/kickstart provisioning

Git workflows

Prometheus

Grafana

tcpdump

Wireshark

Corvil/Beeks latency monitoring

Job description

7+ Years

Full-Time

About the Role

We're looking for a Senior Network Engineer to own the design, build, and operation of our trading network and server estate. You'll be accountable for end-to-end latency across colocation sites, exchange connectivity, and internal fabric - measured in nanoseconds, not milliseconds. This is a hands-on, high-ownership role at the intersection of networking, systems engineering, and trading infrastructure.

What You'll Do

• Architect, deploy, and tune low-latency switching fabric (Arista 7130/7150, Exablaze, Cisco Nexus) and exchange cross-connects
• Manage exchange and market-data connectivity: multicast feeds, A/B feed arbitration, order entry sessions, and colo cage buildouts
• Build and tune trading servers end to end - BIOS/UEFI settings, C-states, P-states, turbo, hyperthreading, NUMA topology, and PCIe lane placement
• Own Linux kernel tuning: IRQ affinity, CPU isolation and shielding, hugepages, scheduler and NIC driver tuning, and kernel bypass (Solarflare Onload, Exanic, DPDK)
• Automate server provisioning, configuration, and deployment across colo sites - no snowflake boxes, everything reproducible from code
• Own the monitoring and observability stack for network, servers, and latency telemetry, including alerting and on-call escalation
• Deploy and maintain precision timing infrastructure: PTP/White Rabbit, GPS grandmasters, hardware timestamping
• Build latency measurement pipelines - tap aggregation, packet capture, and correlation of wire-to-wire and tick-to-trade metrics
• Lead capacity planning, vendor and carrier management, and new venue onboarding
• Mentor and grow a small team of network and systems engineers; set standards and run change control for production trading systems

Required Qualifications

• 7+ years in network/systems engineering, including 3+ years in HFT, prop trading, market making, or a comparable ultra-low-latency environment
• Deep hands-on expertise with low-latency switching, multicast, BGP/OSPF, and exchange connectivity
• Strong Linux internals knowledge - kernel tuning, performance profiling, and troubleshooting under production load
• Demonstrated experience with server-level hardware and firmware optimization for latency
• Configuration management and deployment at scale: Ansible and Puppet (or Salt/Chef), PXE/kickstart provisioning, golden image and firmware lifecycle management, Git-based change workflows, and CI-driven rollout
• Monitoring and observability experience with Prometheus and Grafana, plus practical exposure to several of: Zabbix, Nagios/Icinga, Telegraf/InfluxDB, Netdata, collected, ELK/Loki, SNMP-based network monitoring, LibreNMS, and Corvil/Beeks or equivalent latency monitoring platforms - must be able to build dashboards and alerting from scratch, not just consume them
• Comfort with packet capture and analysis tools (tcpdump, Wireshark, hardware-timestamped capture)
• A track record operating production systems where downtime has direct financial impact

Nice to Have

• Python (or C/C++, Go) for automation, telemetry pipelines, and latency analysis tooling
• FPGA/SmartNIC exposure (Xilinx, Napatech, Arista MetaWatch)
• Terraform, containerization, and infrastructure-as-code beyond config management
• Familiarity with venue-specific protocols (FIX, ITCH/OUCH, exchange-native binary APIs)