BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//HasGeek//NONSGML Funnel//EN
DESCRIPTION:Real systems. Real engineers. Real lessons.
X-WR-CALDESC:Real systems. Real engineers. Real lessons.
NAME:Platform Engineering meet-up - Mar 7
X-WR-CALNAME:Platform Engineering meet-up - Mar 7
REFRESH-INTERVAL;VALUE=DURATION:PT12H
SUMMARY:Platform Engineering meet-up - Mar 7
TIMEZONE-ID:Asia/Kolkata
X-PUBLISHED-TTL:PT12H
X-WR-TIMEZONE:Asia/Kolkata
BEGIN:VEVENT
SUMMARY:Check-in at Quintype
DTSTART:20260307T092000Z
DTEND:20260307T093000Z
DTSTAMP:20260414T224052Z
UID:session/PL2i8TXcmkcmtu6MdZDzPu@hasgeek.com
SEQUENCE:2
CREATED:20260305T075458Z
LAST-MODIFIED:20260305T075503Z
LOCATION:Bangalore
ORGANIZER;CN=Rootconf:MAILTO:no-reply@hasgeek.com
BEGIN:VALARM
ACTION:display
DESCRIPTION:Check-in at Quintype in 5 minutes
TRIGGER:-PT5M
END:VALARM
END:VEVENT
BEGIN:VEVENT
SUMMARY:Introduction to the meet-up
DTSTART:20260307T093000Z
DTEND:20260307T093500Z
DTSTAMP:20260414T224052Z
UID:session/XuHrtswUs6GpJTnrHK3eE8@hasgeek.com
SEQUENCE:0
CREATED:20260305T075448Z
LAST-MODIFIED:20260305T075448Z
LOCATION:Bangalore
ORGANIZER;CN=Rootconf:MAILTO:no-reply@hasgeek.com
BEGIN:VALARM
ACTION:display
DESCRIPTION:Introduction to the meet-up in 5 minutes
TRIGGER:-PT5M
END:VALARM
END:VEVENT
BEGIN:VEVENT
SUMMARY:Worker controller at Zomato: scaling Kafka & SQS consumption with 
 a multi-tenant proxy
DTSTART:20260307T093500Z
DTEND:20260307T101000Z
DTSTAMP:20260414T224052Z
UID:session/GbJEmUMFn1CGguLhtPp7a7@hasgeek.com
SEQUENCE:3
CATEGORIES:Talk (30 mins),Elevator pitch done
CREATED:20260305T075515Z
DESCRIPTION:**Session Description**\n\nAt Zomato\, the journey from a cust
 omer placing an order to its fulfillment\, and beyond\, depends on a compl
 ex web of systems running seamlessly in the background. To power this at a
  massive scale\, we rely on an event-driven architecture built on Kafka an
 d SQS. With thousands of topics and queues in production — and new ones 
 being added almost daily as teams roll out features or system improvements
  — managing this ecosystem is no small task. Every new topic or queue de
 mands corresponding compute resources and workers to consume them. Managin
 g these workers consistently\, reliably\, and safely has always been a cha
 llenge — until now. In this talk\, we introduce Worker Controller\, our 
 central multi-tenant standardized consumer and remote processor designed t
 o simplify worker creation\, improve reliability\, allow canary support\, 
 and accelerate feature delivery across the organization.\n\nWorker Control
 ler abstracts away the complexities of queue consumption from Kafka and SQ
 S by shifting the consumer architecture from a traditional pull-based mode
 l to a centralized push-based proxy. Today\, it manages over 900 Kafka top
 ics\, over 100 SQS queues\, and more than 1\,000 consumers\, handling peak
 s of  ~5 million requests per minute. This proxy layer takes ownership of 
 partition-level offset management\, retries\, dead-letter queues\, batchin
 g\, and observability by default. Because consumption and processing are s
 eparated — with the controller centrally consuming and dispatching work 
 to workers — it becomes much easier to handle poison payloads safely. It
 s RPC-driven design makes creating a new worker as simple as writing an RP
 C method\, drastically reducing operational overhead for SREs. More import
 antly\, the push-based model unlocks canary deployments for workers\, allo
 wing developers to test new code in production with controlled traffic\, c
 atch regressions early\, and deploy with confidence. Also being a central 
 component\, Worker Controller simplifies governance\, making it easier to 
 plan for peak days and ensure that all teams follow best practices consist
 ently.\n\n**Key Takeaways**\n\nWe'll explore how Worker Controller address
 es three critical challenges: resilience\, by enforcing retry budgets and 
 DLQs to prevent cascading failures\, handling poison payloads gracefully\;
  control\, by allowing dynamic throttling\, pause/resume\, and gradual rat
 e-limiting of consumption to protect upstream systems\; and fairness\, by 
 ensuring balanced resource allocation across queues so no "noisy neighbour
 " can dominate. In Kafka use cases where strict ordering is not required\,
  Worker Controller also supports parallel consumption from a partition by 
 sacrificing ordering — delivering higher throughput reliably.\n\nBy the 
 end of this talk\, you’ll learn best practices and patterns for building
  reliable and manageable consumers in an event-driven architecture. We’l
 l explore how to balance developer velocity with operational safety\, reth
 ink worker lifecycle management\, and see how centralizing these concerns 
 can unlock both innovation and stability at scale.\n\n\n**Target Audience*
 *\n\nThis session will be beneficial for a wide range of audiences — fro
 m beginners who are just starting to understand event-driven systems\, to 
 experienced engineers and architects looking for ways to scale and manage 
 complex consumer ecosystems.\n\n**Bio**\n\nWe have two speakers for this s
 ession — their bios are shared below.\n\nSakib Malik is a Senior Softwar
 e Engineer (SDE III) at Zomato\, part of the Site Reliability Engineering 
 team building platform services for large-scale\, distributed systems. He 
 has worked on high-performance libraries and tools\, including contributio
 ns to the Golang runtime\, profiler\, and Redis clients. Sakib also design
 ed resilient infrastructure components such as the Worker Controller\, a K
 afka consumer proxy\, and gomemlimit\, a memory limiter for Go services th
 at significantly reduced operational costs. He is passionate about perform
 ance optimization\, reliability\, and building developer tooling that enab
 les teams to move faster with confidence.\n\nDev Kulkarni is a Software En
 gineer (SDE I) at Zomato\, part of the Site Reliability Engineering team. 
 He has contributed to the development of Worker Controller\, Zomato's cent
 ralized Kafka and SQS consumer platform. Dev is interested in building and
  learning about highly scalable\, resilient systems that improve reliabili
 ty and empower teams to deliver at scale.
LAST-MODIFIED:20260305T075557Z
LOCATION:Bangalore
ORGANIZER;CN=Rootconf:MAILTO:no-reply@hasgeek.com
URL:https://hasgeek.com/rootconf/platform-engineering-meet-up-mar-7/schedu
 le/worker-controller-a-multi-tenant-consumer-proxy-for-consumption-at-scal
 e-GbJEmUMFn1CGguLhtPp7a7
BEGIN:VALARM
ACTION:display
DESCRIPTION:Worker controller at Zomato: scaling Kafka & SQS consumption w
 ith a multi-tenant proxy in 5 minutes
TRIGGER:-PT5M
END:VALARM
END:VEVENT
BEGIN:VEVENT
SUMMARY:Demo: Stateful Kubernetes without fear: disaster recovery simplifi
 ed
DTSTART:20260307T101500Z
DTEND:20260307T103500Z
DTSTAMP:20260414T224052Z
UID:session/2mzQQ419F7VRBqe53BSDYX@hasgeek.com
SEQUENCE:6
CATEGORIES:Vendor pitch - dev tool/solution (15 mins)
CREATED:20260305T075604Z
DESCRIPTION:## Abstract\nDisaster recovery (DR) is no longer optional—es
 pecially when dealing with stateful workloads in Kubernetes environments. 
 In this talk\, I’ll dive into how Portworx DR makes it easier to build a
  resilient infrastructure by enabling efficient disaster recovery strategi
 es tailored for Kubernetes.\n\nI’ll start by laying the foundation: why 
 DR is important in today’s cloud-native world\, what RPO (Recovery Point
  Objective) and RTO (Recovery Time Objective) mean in practical terms\, an
 d how they influence our recovery planning.\n\nWe’ll then explore the di
 fferent DR models supported by Portworx—asynchronous and synchronous (Me
 tro DR)—and how each fits specific business needs. I’ll also walk thro
 ugh the mechanisms behind failover and failback\, and how Portworx automat
 es and streamlines these processes. To bring everything together\, I’ll 
 demonstrate a real-world failover and failback scenario to show just how s
 eamless Portworx DR can be in action.\n\n## Takeaways\n* Understand the fu
 ndamentals of disaster recovery\, why it's critical\, and how to implement
  it in Kubernetes environments\n* Gain insight into DR strategies\, best p
 ractices\, and how to approach DR planning for your own clusters\n\n## Tar
 get Audience\n* Software Engineers seeking a practical introduction to dis
 aster recovery concepts and tooling\n* Infrastructure engineers managing K
 ubernetes environments\n\n## Bio\nHi Rootconf! 👋 I'm Siddhant\, current
 ly working at Portworx (Pure Storage) via Infracloud Technologies. I'm par
 t of the Disaster Recovery team\, where we focus on building solutions tha
 t help businesses stay resilient in the face of unexpected failures.\n\nPr
 ev: Platform Engineering team @ DeepSource.
LAST-MODIFIED:20260305T075646Z
LOCATION:Bangalore
ORGANIZER;CN=Rootconf:MAILTO:no-reply@hasgeek.com
URL:https://hasgeek.com/rootconf/platform-engineering-meet-up-mar-7/schedu
 le/disaster-recovery-for-stateful-k8s-workloads-with-portworx-dr-2mzQQ419F
 7VRBqe53BSDYX
BEGIN:VALARM
ACTION:display
DESCRIPTION:Demo: Stateful Kubernetes without fear: disaster recovery simp
 lified in 5 minutes
TRIGGER:-PT5M
END:VALARM
END:VEVENT
BEGIN:VEVENT
SUMMARY:No more alert fatigue: a platform engineer's guide to alerting in 
 Kubernetes
DTSTART:20260307T104000Z
DTEND:20260307T111500Z
DTSTAMP:20260414T224052Z
UID:session/Ebvb5iYHGWLBS5PvbLJGL5@hasgeek.com
SEQUENCE:3
CATEGORIES:Talk (30 mins)
CREATED:20260305T075657Z
DESCRIPTION:## Overview\nAs our Kubernetes footprint grew\, so did the cha
 llenge of understanding what was happening inside our cluster. We found ou
 rselves drowning in logs from multiple sources\, missing critical alerts\,
  and struggling to connect the dots between metrics\, logs\, and events wh
 en things went wrong.\n\nThis talk shares our journey of building a single
  pane for observability. I'll walk you through what worked\, what didn't\,
  and the hard lessons we learned along the way.\n\nWe'll cover:\n\n- How w
 e consolidated logs\, metrics\, and events from multiple sources\n- Buildi
 ng real-time alerting using Clutch\, Temporal\, and VictoriaMetrics\n- Usi
 ng NATs and TimescaleDB for incident response and on-call alerting\n\n## K
 ey takeaways\n- Real-world patterns for consolidating observability data f
 rom multiple sources\n- Common pitfalls to avoid when building a single pa
 ne for alerting\n\n## Audience\nThis talk is for Platform Engineers and In
 frastructure Engineers who are either:\n\n- Setting up observability for a
  new Kubernetes cluster\n- Struggling with fragmented monitoring across mu
 ltiple sources\n- Looking for practical ways to reduce alert fatigue and i
 mprove incident response\n\n## About Me\nI work as a Senior Member of Tech
 nical Staff @ Nutanix Technologies India Pvt Ltd. I have expertise in mana
 ging and maintaining distributed systems\, including Kubernetes\, Victoria
 Metrics\, and HashiStack\, by stitching together different tools to create
  a robust infrastructure. I enjoy tinkering with distributed systems. Outs
 ide of work\, I am an avid reader and enjoy anything related to science fi
 ction.
LAST-MODIFIED:20260305T075728Z
LOCATION:Bangalore
ORGANIZER;CN=Rootconf:MAILTO:no-reply@hasgeek.com
URL:https://hasgeek.com/rootconf/platform-engineering-meet-up-mar-7/schedu
 le/no-more-alert-fatigue-a-platform-engineers-guide-to-alerting-in-kuberne
 tes-Ebvb5iYHGWLBS5PvbLJGL5
BEGIN:VALARM
ACTION:display
DESCRIPTION:No more alert fatigue: a platform engineer's guide to alerting
  in Kubernetes in 5 minutes
TRIGGER:-PT5M
END:VALARM
END:VEVENT
BEGIN:VEVENT
SUMMARY:Conclusion and feedback
DTSTART:20260307T111500Z
DTEND:20260307T112500Z
DTSTAMP:20260414T224052Z
UID:session/9hfogMzKgtYv19kY18mrMb@hasgeek.com
SEQUENCE:1
CREATED:20260305T075807Z
LAST-MODIFIED:20260305T075809Z
LOCATION:Bangalore
ORGANIZER;CN=Rootconf:MAILTO:no-reply@hasgeek.com
BEGIN:VALARM
ACTION:display
DESCRIPTION:Conclusion and feedback in 5 minutes
TRIGGER:-PT5M
END:VALARM
END:VEVENT
BEGIN:VEVENT
SUMMARY:Group photo\, networking and snacks
DTSTART:20260307T112500Z
DTEND:20260307T121000Z
DTSTAMP:20260414T224052Z
UID:session/31wT1HDX6e16d6X6Myj1Li@hasgeek.com
SEQUENCE:2
CREATED:20260305T075741Z
LAST-MODIFIED:20260305T075753Z
LOCATION:Bangalore
ORGANIZER;CN=Rootconf:MAILTO:no-reply@hasgeek.com
BEGIN:VALARM
ACTION:display
DESCRIPTION:Group photo\, networking and snacks in 5 minutes
TRIGGER:-PT5M
END:VALARM
END:VEVENT
END:VCALENDAR