Srihari Kookal

Engineering Manager | Technical Lead | Distributed Systems | AI Engineering · Frankfurt am Main, Germany

Srihari Kookal.

Github: srih4ri · Linkedin: srih4ri · Staff-engineer version

Summary

Engineering leader with 16+ years of experience building and operating business-critical software systems across SaaS, fintech, insurance and data-intensive platforms.

Currently the reporting manager of a six-engineer data collection team at Omnia Retail. Previously led Scripbox’s fifteen-member product team as AVP Technology, winning the company’s Best Manager Award, and ran Clark’s Marketing Tech team as de facto engineering manager.

Combines people management, roadmap planning, hiring and mentoring with hands-on depth in distributed systems, cloud infrastructure, reliability and AI-powered engineering workflows.

Core Expertise

  • Leadership & Management: people management, career growth, appraisal systems, roadmaps, OKRs, Agile delivery, hiring, mentoring
  • AI Engineering: AI agents, Claude Code, MCP, RAG, LLM-powered workflows, developer productivity automation
  • Distributed Systems: data pipelines, horizontal scaling, sharding, asynchronous processing, high availability
  • Cloud & Platform Engineering: AWS, Kubernetes, EKS, Terraform, Terragrunt, GitOps, Docker
  • Observability & Reliability: OpenTelemetry, Prometheus, Grafana, Datadog, distributed tracing, SLOs, incident management
  • Backend Engineering: Scala, Python, Ruby, Ruby on Rails, Go, PostgreSQL, Redis, Elasticsearch
  • Languages: English (fluent), German (A1)

Experience

Team Lead, Data Collection, Omnia Retail Technical Leadership + People Management

Sep, 2025 - Present

Leads the six-engineer team behind the Scala/Python market-data pipeline on Kubernetes; built the team's AI-agent workspace and LLM alert-investigation toolkit, and moved the largest workload class to ARM64.

Lead Team Daedalus, the six-engineer Market Data collection team behind Omnia Retail’s pricing-intelligence product for retailers, as their reporting manager. The team owns a large Scala/Python data pipeline running on Kubernetes.

Leadership & AI Engineering
  • Contribute to hiring, coding principles and onboarding documentation.
  • Built a shared AI-agent workspace for the team, wiring Claude Code into Jira, Confluence, log search, dashboards, the data warehouse and internal databases through MCP servers, with reusable skills for incident investigation, dashboards, alerts and pull-request review.
  • Built an alert-investigation toolkit that combines log search, VictoriaMetrics, kubectl and an LLM-driven agent, with trend awareness, alert fingerprinting, contract-impact tracking mapped to product KPIs, and threaded Slack notifications.
  • First responder for production incidents on the pipeline, owning root-cause analysis, follow-up fixes and customer-facing communication.
Data Platform & Infrastructure
  • Drove the migration of the cluster’s largest workload class to ARM64/Graviton nodes, adding parallel multi-arch CI/CD pipelines and right-sizing Kubernetes resources across worker types, which recovered significant cluster capacity and reduced compute spend.
  • Operated and hardened a high-throughput data-collection subsystem, designing horizontal sharding, autoscaling, observability, PodDisruptionBudgets and Flux image-update automation around a Redis-backed supporting component.
  • Contributed to migrating the team’s production workloads onto a new Flux GitOps-managed EKS cluster, fixing ingress, IPv6 dual-stack, Karpenter and KEDA issues during cutover.
  • Added OpenTelemetry tracing across the Scala services, with a local Jaeger/OTel-collector setup for testing.
  • Authored Python diagnostic tooling for data-freshness and routing analyses, and drove the team’s Grafana migration from AWS to Azure Managed Grafana with Azure AD.

Technologies: Scala, Python, Kubernetes, AWS EKS, Karpenter, KEDA, Flux, Terraform, Docker, ARM64, Redis, PostgreSQL, TimescaleDB, MongoDB, Elasticsearch, OpenTelemetry, Jaeger, VictoriaMetrics, Prometheus, Grafana, Databricks, Claude Code, MCP.

Staff Engineer, Clark Technical Leadership

Feb, 2023 - Aug, 2025

Kept the Rails monolith behind 90% of Clark's backend at 100% uptime through a prime-time TV event; cut primary database load by 30%, cloud cost by 10%, errors by 60% and alarms by 70%.

Maintained the 10-year-old Ruby on Rails monolith that ran 90% of Clark’s backend services, and planned and executed its Ruby and Rails upgrades to meet software supply chain standards.

  • Collaborated with engineering managers and the director to set and achieve annual engineering goals for Clark’s German business division.
  • Maintained the Domain-Driven Design architecture, reviewed solution designs for all major backend changes, and mentored engineers as code owner of critical components.
  • Led the migration from Delayed Job to Sidekiq, coordinating a cross-team effort to move close to 500 background jobs and improving background job performance by 20%.
  • Proposed and built an elastically scaling signup system, and planned and executed a quarter of resilience work (k6 load tests; Puma, Nginx and Kubernetes tuning) that kept 100% uptime through a prime-time national TV event viewed by millions across Germany.
  • Set up a read-replica database and wrote custom middleware to route all GET requests to it, reducing primary database load by 30%.
  • Implemented vertical and horizontal pod autoscaling, reducing monthly cloud costs by 10%.
  • Reduced the error rate by 60% and monitoring alarms by 70% by cutting alert noise, and set up a triage process routing new errors to owning teams.
  • First responder for major incidents; established DORA-compliant deployment, rollback and incident-response processes.
  • Carried out zero-downtime upgrades of Clark Germany’s database across major PostgreSQL versions, using AWS Blue/Green deployments to migrate all environments within a week.

Technologies: Ruby on Rails, Go, Python, Kubernetes, AWS (EKS, RDS, Aurora), Terraform, Terragrunt, Helm, PostgreSQL, Redis, Sidekiq, Datadog, Prometheus, Grafana, k6, Gemini, LangChain.

Tech Lead, Marketing Technology, Clark Technical Leadership

Feb, 2021 - Feb, 2023

Ran the Marketing Tech team as de facto engineering manager; cut campaign go-live from two weeks to almost instant and automated voucher payouts.

  • De facto Engineering Manager for the Marketing Tech team; worked with the Marketing team liaison to plan and execute quarterly initiatives and goals.
  • Ran Agile sprint ceremonies and met the company’s lead-time and cycle-time goals.
  • Built an in-house Marketing Campaign Management tool that automated all of Clark’s marketing reward programs, cutting campaign go-live time from 2 weeks to almost instant.
  • Proposed and implemented automated voucher payouts, increasing customer satisfaction and NPS.

Senior Software Engineer, Clark Individual Contributor

Nov, 2020 - Feb, 2021

Built an in-house A/B testing framework for the primary signup funnel and decommissioned legacy codebases.

  • Built an in-house A/B testing framework for backend and frontend applications, assigning variants with a weighted hash algorithm, to improve conversion of the primary signup funnel with marketers.
  • Decommissioned legacy codebases, reducing system complexity.

Associate Vice President - Technology, Scripbox Leadership

Feb, 2020 - Oct, 2020

Led the fifteen-member product team with the founders, CTO and CPO; won the annual Best Manager Award.

Led the fifteen-member product team, working closely with the founders, CTO and CPO.

  • Managed software engineers, quality analysts and business analysts through one-on-ones and career and growth trajectories; won the company’s annual Best Manager Award.
  • Planned and executed the annual vision and quarterly roadmaps with Product, and set up a skill-based performance and appraisal system for engineering with company leadership.
  • Completed the customer data migration after the company’s first major merger and acquisition.
  • Worked with the VP Operations to achieve order-processing KPIs, and launched projects that met the in-house call center’s operational KPIs, helping Customer Success reach a very high NPS.
  • Built portfolio growth calculation software that computed ~2 million data points in under 30 minutes, using Elixir, PostgreSQL and CBLAS matrix computation.

Product Engineer, Scripbox Technical Leadership + Individual Contributor

Feb, 2017 - Feb, 2020

Led the backend team for core services; built the BPM system, portfolio reporting and paperless onboarding that cut time-to-invest from days to hours.

  • Led the backend team for Scripbox’s core services, and onboarded its four Rails developers to React, switching from Handlebars with no impact on delivery times.
  • Built a Business Process Management system from scratch that made Scripbox’s processes measurable, efficient and fast, and a highly concurrent portfolio reporting system in a month using Elixir and PostgreSQL.
  • Built key components of a paperless onboarding process, reducing the time to invest in a mutual fund from days to hours.
  • Rolled out the Rails 4.3 to 5.1 upgrade with near-zero downtime; first responder and on-call for core services and the call center’s uptime.

Earlier roles Individual Contributor

July, 2010 - Dec, 2016

Rails and JavaScript engineering across SaaS, community engagement and education products, including Kerala's statewide school ERP.

  • Software Consultant, Big Binary (July 2015 - Dec 2016): ran deployments of a Rails 2.3 app for 200+ clients and worked on its multi-tenant SaaS rewrite.
  • Software Architect, Bang The Table (Jan 2013 - July 2015): built two community engagement products in Rails and JavaScript.
  • Software Engineer, Foradian Technologies (July 2010 - May 2012): built Sampoorna, the school ERP used in all of Kerala’s public schools, processing data for 400k new students every year; worked on Fedena, the open-source school management system.

Selected Projects

  • XPNS (xpns.fun): personal finance app for expenses, income and investments; Ruby on Rails, EmberJS and PostgreSQL on Cloudflare and Hetzner. Blog post about building XPNS
  • Foaps (consulting): built backend systems for orders, payments and customer support (Rails, PostgreSQL, Redis), including a WhatsApp food delivery system; cut report load times from 10 minutes to under 5 seconds.
  • Debian Ruby (open source): packaged Ruby gems for Debian with the debian-ruby team for the Diaspora project.

Education

BTech, Electronics and Communication, Lal Bahadur Sasthri College of Engineering, July 2010.