Senior Software Engineer
I build distributed systems and
multi-cloud infrastructure.
Hi, I’m Gourish. From GPU orchestration for LLMs to multi-cluster Kubernetes networking — I engineer systems that scale, route, and optimize at the highest level.
Expertise
What I Build
Systems that scale across clusters, clouds, and constraints.
Distributed Systems
Building highly available, fault-tolerant architectures across complex network boundaries.
GPU / AI Infrastructure
Deep expertise in Kubernetes, containers, and building production-grade platforms.
Cloud-Native Infrastructure
Orchestrating dynamic GPU allocation and scaling for heavy LLM workloads.
Performance & Cost Optimization
Engineering predictive autoscaling that slashes cloud bills without degrading performance.
Featured Work
Things I've Built
Production-grade systems, not side projects.
CNCF Sandbox
KubeSlice
Multi-cluster Kubernetes networking platform. Turns complex distributed infra into a composable, secure overlay network — now an open-source ecosystem under CNCF.
→ NSM-based overlay networking across clusters
→ Mutual authentication and end-to-end security
→ Led the open-sourcing and CNCF Sandbox submission
→ Built for enterprise multi-tenancy at scale
Impact
Turned complex infra into a thriving open-source ecosystem
Kubernetes
NSM
Go
CNCF
eBPF
LLM Infrastructure
Elastic GPU Service
Dynamic GPU allocation platform designed for LLM workloads. Supports NVIDIA & AMD, tracks real-time utilization, and makes GPU infrastructure as elastic as compute.
→ Dynamic GPU allocation per workload demand
→ NVIDIA & AMD GPU support
→ Real-time utilization telemetry
→ Scheduler integration for intelligent placement
Impact
Made GPU infrastructure elastic and observable for LLM workloads
Kubernetes
NVIDIA
AMD
Go
Prometheus
Cost Engineering
Smart Scaler
Predictive autoscaling platform for multi-cloud deployments. Reduced cloud spend by 20–70% without sacrificing performance or availability.
→ Predictive scaling ahead of traffic spikes
→ Multi-cloud support across AWS, GCP, Azure
→ Custom metrics and policy engine
→ Zero-downtime scaling events
Impact
20–70% cloud cost reduction across production deployments
Kubernetes
HPA
KEDA
Go
Multi-cloud
Networking
Smart Traffic Director
DNS-based multi-cloud traffic management system. Routes requests intelligently across cloud providers with high availability and minimal latency.
→ DNS-based routing with health-aware failover
→ Multi-cloud traffic distribution policies
→ Latency-optimized request routing
→ Real-time traffic observability
Impact
Eliminated single-cloud lock-in with intelligent global traffic control
DNS
Envoy
Multi-cloud
Go
Istio
Community
Open Source
Building in public. Contributing to the ecosystem.
Philosophy
I don't just build systems — I contribute to the ecosystem.
The best infrastructure is the kind others can build on. Every contribution to open source is infrastructure for the next system someone needs to build.
CNCF Contributor
Active contributor to Cloud Native Computing Foundation projects. Led KubeSlice through the CNCF Sandbox acceptance process — from internal tool to public ecosystem.
LFX Mentor
Linux Foundation LFX Mentor — helping the next generation of systems engineers contribute meaningfully to production-grade open source projects.
KubeSlice — CNCF Sandbox
Turned an internal multi-cluster networking platform into an accepted CNCF Sandbox project. Open source from architecture through documentation
Timeline
Experience
Where the systems were built.
2021 — Present
Senior Software Engineer
Avesha Systems
Built KubeSlice (CNCF Sandbox), Elastic GPU Service, Smart Scaler, and Smart Traffic Director. Led open-sourcing efforts, contributed to CNCF ecosystem, and shipped production infrastructure used across enterprise deployments.
2019 — 2021
Software Engineer
Citrix
Built and maintained cloud-native networking and infrastructure components. Worked on distributed systems and Kubernetes-based platforms serving enterprise clients at scale.
Writing
Thoughts on Systems & Scale
Notes from building distributed systems in production.
Get In Touch
Let's build something hard.
If you're working on complex infrastructure — distributed, cloud-native, or AI-scale — let's talk.
biradar.gourish@gmail.com
© 2026 Gourish Biradar
Built with precision. Deployed with care.