Anmol Chawla

Co-Founder

Bengaluru, Karnataka, India8 yrs 2 mos experience

AI ML PractitionerAI Enabled

Key Highlights

Expert in Kubernetes and distributed systems.
Proven track record in optimizing CI/CD processes.
Skilled in implementing AI solutions for real-time applications.

Stackforce AI infers this person is a SaaS and Logistics expert with strong DevOps capabilities.

Contact

Skills

Core Skills

KubernetesDevopsDatabasesPythonAiMobile Application DevelopmentProduct DevelopmentCi/cdSoftware DevelopmentInfrastructure

Other Skills

AWSAzureCC++Continuous Integration and Continuous Delivery (CI/CD)DartFastAPIFlaskFlutterGitGitHub ActionsGoogle CloudKotlinKotlin CoroutinesLinux

About

An experienced senior software engineer working with a demonstrated history of working with software products. Skilled in working with distributed systems as well as web and mobile technologies. Familiar with modern ML inference pipelines and scaling them for throughput.

Experience

Dashtoon

Founding Engineer

Jan 2023 – Present · 3 yrs 2 mos · Bengaluru, Karnataka, India · On-site

Streamlined content management on platform by migrating from single-region replicated
Postgres database to multi-region geo-replicated Azure Cosmos (No-SQL) database, creating
a single source of truth for content and eliminating consistency problems between multiple
databases. Was achieved with 0 downtime by using Dual Read Write strategy, and also
resulted in a 40% reduction in latencies
Scaled reader service to service requests globally by establishing a multi-region distributed
setup. In the process of which a formal user service was also setup, handling assigning users
to a region and re-routing requests transparently. Allowed establishing a point of presence
in different regions and isolating financial records to the user’s region. The setup involved
usage of the Azure’s global load-balancer and resulted in an 81% reduction in latencies globally
Created and managed 7 production Kubernetes clusters spread across Azure AKS, Google
Cloud GKE and AWS EKS, granting access to varied GPU nodes and handling traffic across
regions.
Setup and maintained APM tooling via Open Telemetry and Signoz across 7 Kubernetes
clusters, giving real time insights on latencies and errors. Reduced mean time to resolve
issues by 50%
Implemented Generative AI pipelines using Diffusers and FastAPI servers for image generation
requirements, powering the core of Dashtoon’s image generation capabilities. Achieved 25
seconds per high resolution image generation time on L4 GPUs
Created an internal advertisement platform on the app for serving advertisements based
on content published on the app. Resulted in an increase in crosswalk by 25%
Setup CI/CD pipelines across the organisation for Kotlin, Python, React and Flutter repositories
using Github Actions, reducing the time spent on deployments and management of micro-
services.
Conceptualised and implemented payments micro-service for handling user transactions
and facilitating internal economy.

LinuxDatabasesFlutterDartInfrastructureMobile Application Development+8

Udaan.com

3 roles

Product Engineer II

Promoted

Jan 2022 – Jan 2023 · 1 yr

Created a platform for tracking last mile deliveries in real time by the operations teams at
last mile hub. Allowed the operations teams to set delivery SLAs and nudge drivers for timely
deliveries. Resulted in a reduction of 20% in overall delivery time for the day. Handled over
1K location updates per second
Migrated entire organisation, with over 200 repositories, from Jenkins to Github Actions
for CI/CD pipelines. Migration was automated by using a Groovy AST parser to convert
Jenkins DSL to YAML. Action runners were setup as self-hosted runners. Resulted in the
setup becoming easier to modify and more approachable. Reduced build times by 50% and
ha costs saving of 40%
Managed over 400 Virtual Machines across 3 Kubernetes clusters and 7 datacenters with a
mix of Azure and on-premise.
Setup Istio proxy across clusters to facilitate inter-cluster communication between services

Continuous Integration and Continuous Delivery (CI/CD)InfrastructureProduct DevelopmentSoftware DevelopmentKotlinKotlin Coroutines+3

Product Engineer

Jul 2020 – Feb 2022 · 1 yr 7 mos

Facilitated reduction in organisational infrastructure cost by implmenting Horizontal Pod
Autoscaler across production clusters. Teams were asked to reduce resource requests and
instances required. Overall effort resulted in a reduction of 80% of infrastructure costs
Established a virtual warehousing concept for facilitating just-in-time deliveries from vendor
to the buyer, without requiring item to be inwarded at a physical warehouse. Allowed savings
of upto a day in delivery timelines
Operationalized internal routing of deliveries which used pigeon hole and k-means clustering
for grouping of delivery locations and historical analysis for the route to be taken within a
cluster. Automated the burden of manual routing on the last mile hub team and allowed for
efficiencies to be made in the delivery routes. Reduced the process from 2 hours of manual
effort to 15 minutes of automated routing