Abhinav Pratap profile photo 1 on Tal

Abhinav Pratap

DevOps Intern @ Katonic AI | Kubernetes & Docker | Cloud Infrastructure | AI Infra

Delhi

1+ year of work experience

Katonic AI
Jul 202510 mosPresent

Mumbai, Maharashtra, India

• Deployed and managed 10+ applications on Kubernetes using Deployments, Services, and Istio Gateways/VirtualServices, while automating CI/CD with GitHub Actions for faster, consistent rollouts. • Deployed and tested Harbor as a self-hosted registry and mirrored organization images from Quay to support sovereign and air-gapped deployment environments. • Maintained and upgraded developer workspace environments (RStudio, VS Code, Jupyter) by managing Kubernetes workloads and container images, ensuring stable development environments for platform users. • Resolved container runtime compatibility issues with Red Hat OpenShift under restricted SCC policies by modifying Dockerfiles to run as non-root users and rebuilding selected images on Red Hat UBI as part of the platform’s OpenShift certification process. • Optimized large Docker images by restructuring layers and dependencies, reducing image size by 35–50%, cutting reported vulnerabilities by∼90%, and improving build efficiency.
Hooman Digital
Apr 20254 mosJul 2025

Kolkata, West Bengal, India

• Built reusable containerized LLM serving templates using Docker and shell scripts for deployment on Nosana (a decentralized GPU platform similar to RunPod). • Implemented dynamic model fetching from Hugging Face with runtime loading via vLLM and Ollama. • Diagnosed and resolved GPU worker container failures by validating GPU compatibility and fixing runtime configurations.
PeerHub
Dec 20245 mosApr 2025

Noida, Uttar Pradesh, India

• Built and managed AWS infrastructure using Docker and GitHub Actions for automated builds and deployments. • Designed VPC architecture with public/private subnets, route tables, and security groups to ensure secure service connectivity. • Configured ALB and Auto Scaling Groups to distribute traffic and scale EC2 instances during traffic spikes. • Implemented Amazon CloudWatch monitoring and alerting with custom metrics to track system performance.

Abhinav’s AI Stack

cursor
cursor
github copilot
github copilot
ollama
ollama
openai
openai

Education

kiet group of institutions

Bachelor of Technology - BTech, Information Technology

kiet group of institutions

20224 yrs2026

Join Abhinav and 10k+ other professionals on

Tal Boss

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

Download Tal Boss

Abhinav Pratap on Tal

Headline

DevOps Intern @ Katonic AI | Kubernetes & Docker | Cloud Infrastructure | AI Infra

About

I work on infrastructure and enjoy figuring out how to make systems run smoothly. Recently, I’ve been exploring AI infrastructure, model deployments, and different ways to run AI workloads more efficiently. I also share what I learn on my blog whenever I experiment with something interesting. In the meantime, I enjoy traveling and cooking.

Location

Delhi

AI stack

cursor

github copilot

ollama

openai

Work experience

DevOps Intern at Katonic AI, Jul 2025 - Present · 10 mos, Mumbai, Maharashtra, India, • Deployed and managed 10+ applications on Kubernetes using Deployments, Services, and Istio Gateways/VirtualServices, while automating CI/CD with GitHub Actions for faster, consistent rollouts. • Deployed and tested Harbor as a self-hosted registry and mirrored organization images from Quay to support sovereign and air-gapped deployment environments. • Maintained and upgraded developer workspace environments (RStudio, VS Code, Jupyter) by managing Kubernetes workloads and container images, ensuring stable development environments for platform users. • Resolved container runtime compatibility issues with Red Hat OpenShift under restricted SCC policies by modifying Dockerfiles to run as non-root users and rebuilding selected images on Red Hat UBI as part of the platform’s OpenShift certification process. • Optimized large Docker images by restructuring layers and dependencies, reducing image size by 35–50%, cutting reported vulnerabilities by∼90%, and improving build efficiency.

Backend and DevOps Intern at Hooman Digital, Apr 2025 - Jul 2025 · 4 mos, Kolkata, West Bengal, India, • Built reusable containerized LLM serving templates using Docker and shell scripts for deployment on Nosana (a decentralized GPU platform similar to RunPod). • Implemented dynamic model fetching from Hugging Face with runtime loading via vLLM and Ollama. • Diagnosed and resolved GPU worker container failures by validating GPU compatibility and fixing runtime configurations.

DevOps and Cloud Intern at PeerHub, Dec 2024 - Apr 2025 · 5 mos, Noida, Uttar Pradesh, India, • Built and managed AWS infrastructure using Docker and GitHub Actions for automated builds and deployments. • Designed VPC architecture with public/private subnets, route tables, and security groups to ensure secure service connectivity. • Configured ALB and Auto Scaling Groups to distribute traffic and scale EC2 instances during traffic spikes. • Implemented Amazon CloudWatch monitoring and alerting with custom metrics to track system performance.

Education

Bachelor of Technology - BTech, Information Technology, kiet group of institutions

Abhinav Pratap - DevOps Intern in Delhi | Tal