Suryanshu Gupta profile photo 1 on Tal

Suryanshu Gupta

Full-Stack Engineer | Go, Kubernetes, AWS | Building GPU-as-a-Service Infrastructure for AI/ML Workloads @ NAVA

Bengaluru

1+ year of work experience

Nava
Jul 20251 yr 2 mosAug 2026

Bengaluru, Karnataka, India

- Nava is building a GPU‑as‑a‑Service platform on Kubernetes for AI workloads. I own core backend services, infrastructure, and authentication.​ - Architected GPUaaS on Kubernetes using KubeVirt with automated VM lifecycle + persistent volumes, turning manual GPU provisioning from hours into minutes.​ - Built KubeVirt installers/validators to standardize GPU VM creation and storage attachment, cutting environment setup errors and support load.​ - Designed centralized AuthN/AuthZ (sessions, roles, OAuth2/JWT) for multi‑service backend, simplifying tenant onboarding and tightening access control.​ - Shipped CI/CD pipelines targeting Tanzu Kubernetes clusters for banking customers, enabling one‑click, repeatable deployments with safe rollbacks.​ - Implemented SSO tenant onboarding and session‑limit policies (max concurrent sessions, auto‑kill oldest), reducing friction for bulk signups and improving security. - Led 700TB+ data migration from AWS to in-house storage infrastructure, designing parallelized transfer pipelines, integrity validation, and rollback strategies to ensure zero data loss and minimal downtime. - Owned and shipped end-to-end self-service signup flow, including onboarding, tenant provisioning, auth integration, and resource allocation, enabling users to go from signup → GPU instance in minutes without manual intervention.
aarna.ml
Jan 20257 mosJul 2025

Bengaluru, Karnataka, India

- Built Go microservices and PostgreSQL‑backed APIs for a GPUaaS platform, designed for high concurrency and horizontal scaling.​ - Optimized API and query performance, reducing average response time by ~35% and increasing throughput for GPU job workflows.​ - Co‑developed CI/CD pipelines (Kubernetes, Ansible, Makefiles) to standardize environment provisioning and rollouts.​ - Improved observability with structured logging, metrics, and health checks, cutting incident debug time and making on‑call easier.​ - Integrated OAuth2/JWT‑based AuthN/AuthZ to secure multi‑tenant services.​
aarna.ml
Sep 202411 mosJul 2025

aarna.ml
Sep 20245 mosJan 2025

Bengaluru, Karnataka, India

GitHub
@suryanshu185

Suryanshu’s AI Stack

claude code
claude code
codex
codex
kiro
kiro
githubcopilot
githubcopilot
LiteLLM
LiteLLM
vLLM
vLLM

Education

Thapar Institute of Engineering & Technology

Computer science and business systems, Computer Science

Thapar Institute of Engineering & Technology

20214 yrs2025
Thapar Institute of Engineering & Technology

Bachelor of Engineering, Computational Science

Thapar Institute of Engineering & Technology

Join Suryanshu and 10k+ other professionals on

Tal Boss

Tal is the app where bosses meet engineers & builders directly, without middlemen

Download Tal Boss

Suryanshu Gupta on Tal

Headline

Full-Stack Engineer | Go, Kubernetes, AWS | Building GPU-as-a-Service Infrastructure for AI/ML Workloads @ NAVA

About

Yo! You've just landed on my page. Good, now that you're here, you should know I'm a full stack engineer who lives where code meets infrastructure. Right now that means building GPU-as-a-Service (GPUaaS) platforms on Kubernetes that power AI/ML and GenAI compute workloads Go microservices, KubeVirt-based VM lifecycle automation, CI/CD pipelines, and AWS infra, with a track record of shipping measurable gains in reliability, deployment speed, and API performance across NAVA, Aarna Networks, and Rivet Technology. Throw anything my way — backend, DevOps, or the AI infra in between and I'll figure it out. If you think our interests match, let's connect and build something crazy good.

Location

Bengaluru

AI stack

claude code

codex

kiro

githubcopilot

LiteLLM

vLLM

Work experience

Founding Software Engineer at Nava, Jul 2025 - Aug 2026 · 1 yr 2 mos, Bengaluru, Karnataka, India, - Nava is building a GPU‑as‑a‑Service platform on Kubernetes for AI workloads. I own core backend services, infrastructure, and authentication.​ - Architected GPUaaS on Kubernetes using KubeVirt with automated VM lifecycle + persistent volumes, turning manual GPU provisioning from hours into minutes.​ - Built KubeVirt installers/validators to standardize GPU VM creation and storage attachment, cutting environment setup errors and support load.​ - Designed centralized AuthN/AuthZ (sessions, roles, OAuth2/JWT) for multi‑service backend, simplifying tenant onboarding and tightening access control.​ - Shipped CI/CD pipelines targeting Tanzu Kubernetes clusters for banking customers, enabling one‑click, repeatable deployments with safe rollbacks.​ - Implemented SSO tenant onboarding and session‑limit policies (max concurrent sessions, auto‑kill oldest), reducing friction for bulk signups and improving security. - Led 700TB+ data migration from AWS to in-house storage infrastructure, designing parallelized transfer pipelines, integrity validation, and rollback strategies to ensure zero data loss and minimal downtime. - Owned and shipped end-to-end self-service signup flow, including onboarding, tenant provisioning, auth integration, and resource allocation, enabling users to go from signup → GPU instance in minutes without manual intervention.

Software Engineer Intern at aarna.ml, Jan 2025 - Jul 2025 · 7 mos, Bengaluru, Karnataka, India, - Built Go microservices and PostgreSQL‑backed APIs for a GPUaaS platform, designed for high concurrency and horizontal scaling.​ - Optimized API and query performance, reducing average response time by ~35% and increasing throughput for GPU job workflows.​ - Co‑developed CI/CD pipelines (Kubernetes, Ansible, Makefiles) to standardize environment provisioning and rollouts.​ - Improved observability with structured logging, metrics, and health checks, cutting incident debug time and making on‑call easier.​ - Integrated OAuth2/JWT‑based AuthN/AuthZ to secure multi‑tenant services.​

Back End Developer at aarna.ml, Sep 2024 - Jul 2025 · 11 mos

Technical Consultant at aarna.ml, Sep 2024 - Jan 2025 · 5 mos, Bengaluru, Karnataka, India

Education

Computer science and business systems, Computer Science, Thapar Institute of Engineering & Technology

Bachelor of Engineering, Computational Science, Thapar Institute of Engineering & Technology

Public links

github: https://github.com/suryanshu185