Skip to content
View infravibe's full-sized avatar

Block or report infravibe

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
infravibe/README.md

banner

👋 Hi, I'm Akash Sahani

I build systems, not just services — distributed backend, infrastructure & AI working together at scale

🚀 About Me

Software Engineer with experience building large-scale distributed systems, AI platforms, and real-time infrastructure.

  • Built 50+ production-grade microservices powering enterprise systems
  • Designed event-driven architectures using Kafka, Spark & Kubernetes
  • Deployed and scaled LLMs on Kubernetes using Ray & DeepSpeed
  • Developed real-time voice AI systems (WebRTC + SIP + LiveKit)
  • Focused on performance, observability, and production reliability

⚡ What I Work On

  • 🧩 Distributed backend systems (Java, Spring Boot, gRPC)
  • ☁️ Cloud-native infrastructure (AWS EKS, Terraform, Helm)
  • ⚡ Real-time communication systems (WebRTC, LiveKit, SIP)
  • 🤖 AI/LLM systems (LLMOps, RAG, agentic workflows)
  • 📊 Data platforms (Spark, Kafka, Iceberg, Hive)

🧰 Tech Stack

Backend:
Java, Spring Boot, Python (FastAPI, Flask), gRPC

Infrastructure & DevOps:
Kubernetes (EKS), Terraform, Helm, Jenkins, AWS, GCP, Ace Cloud

Realtime Systems:
WebRTC, LiveKit, SIP, STUN/TURN

Data & ML:
Apache Spark, Kafka, Feast, Kubeflow

Observability:
Grafana Stack (Mimir, Loki, Tempo, Alloy)

🚀 Featured Work

🔹 Voice AI Platform

Real-time, low-latency voice system using WebRTC, SIP & LiveKit on AWS EKS

  • Solved NAT traversal & media routing challenges
  • Designed auto-scaling, fault-tolerant infrastructure

🔹 Elevate AI — Multi-Tenant SaaS Platform

Event-driven microservices platform with AI workflows

  • Multi-tenant architecture with strict data isolation
  • Kafka-based async processing for scalability
  • Integrated AI content generation pipelines

🔹 LLMOps Platform

Deployed LLMs on Kubernetes using Ray & DeepSpeed

  • Distributed inference at scale
  • Integrated with enterprise ML pipelines

🔹 MLHub — Data & Feature Platform

End-to-end ML platform with governance & feature store

  • Spark-based pipelines for batch & real-time processing
  • Kubeflow orchestration + Feast feature store
  • 100% data lineage with Apache Atlas

📊 Impact

  • ⚡ Reduced data processing time from 20 mins → 1.5 mins
  • 🚀 Achieved <1s observability alerting across 100+ services
  • 🔁 Automated CI/CD pipelines with zero-error deployments
  • 📈 Built systems supporting large-scale AI-driven workloads

📈 Current Focus

  • MLOps & AI Infrastructure
  • Distributed Systems at Scale
  • Real-time Communication Systems

✍️ Writing

I write about backend systems, DevOps, and AI infrastructure:
👉 https://medium.com/@akashsahani2001

🤝 Connect

  • LinkedIn
  • Portfolio
  • GitHub

Popular repositories Loading

  1. apache apache Public

    Shell 2 1

  2. observability observability Public

  3. observability-examples observability-examples Public

    Python

  4. livekit-aws-eks-infra livekit-aws-eks-infra Public

    Production-grade Terraform infrastructure for deploying LiveKit Server on AWS EKS with STUN/TURN, NLB, and scalable WebRTC networking.

    HCL

  5. livekit-helm livekit-helm Public

    Forked from livekit/livekit-helm

    LiveKit Helm charts

    Go Template

  6. claude-cli-agent claude-cli-agent Public

    Run Claude-powered coding agents safely inside Docker containers.

    Shell