Skip to content

About Me

Hi! my name is Amit Upadhyay, a Backend Engineer passionate about building high-scale distributed systems and exploring the world of AI/ML. Welcome to the Dwar-AI Blog, where I document my technical deep dives, architectural discoveries, and adventures in integrating LLMs into robust backend systems.

Backend Engineer (Central AdTech)

Delivery HeroBerlinJan 2023 — Present

  • Scale & Architecture: Engineered and scaled the core tracking and billing pipeline across the global AdTech platform (Foodpanda, Glovo, Talabat, etc), processing peak loads of 30,000+ events/sec via Go, Kafka, and GCP/AWS on Managed Kubernetes (GKE/EKS).
  • Performance: Drove significant system-wide latency reductions across high-throughput pipelines through low-level design refactoring and async batching.
  • Applied AI: Designed and deployed a production RAG and MCP server ecosystem enabling Slack AI agents to autonomously triage production alerts, cutting incident response time.
  • Observability: Enhanced system observability using Prometheus and Grafana to detect anomalies across high-throughput pipelines in production, preventing revenue bleed and ensuring high availability.
  • Infrastructure: Integrated Atlantis to manage Terraform changes via GitHub PRs and migrated CI pipelines to self-hosted Drone CI.
  • Tech Stack: Go, Python, PostgreSQL, Kafka, Redis, gRPC, Docker, GCP (Pub/Sub, CloudSQL, GKE), AWS (EKS, SQS, SNS), Terraform, Atlantis, Drone CI, Prometheus, Grafana.

Senior Software Engineer

TrivagoDüsseldorfNov 2021 — Nov 2022

  • Migrated legacy services to Go-based microservices as part of a platform modernization effort, improving deployment speed and reducing operational overhead.
  • Transitioned services from request-response to event-driven architecture using Kafka, reducing inter-service coupling and improving system resilience.

Backend Engineer (Payments)

Ola CabsBangaloreAug 2019 — Aug 2021

  • Worked on the payments team at Ola Cabs, one of India’s largest ride-hailing platforms at the time, handling millions of daily ride transactions.
  • Integrated and maintained multiple payment providers (Adyen, PayU, PayPal) supporting UPI, Credit/Debit Cards, Wallets and Cash.
  • Built fraud detection and transaction validation mechanisms to reduce payment failures and fraudulent activity across the platform.

Personal Projects & Full SDLC

Professionally, my entire career has been dedicated to engineering and scaling backend systems. However, through personal projects and hackathons, I have explored the full Software Development Life Cycle (SDLC) by building applications with C++, Node.js, and Flutter. This hands-on experience across the stack allows me to deeply understand end-to-end system architecture and empathize with the client teams consuming my APIs.

Bachelor of Engineering in Computer Science

Army Institute of Technology2014 — 2018

Backend & Cloud

Go, Python, Java, PostgreSQL, Redis, Kafka, gRPC, Docker, Kubernetes (GKE/EKS), Terraform, Atlantis, Drone CI, GCP, AWS.

Observability & Messaging

Prometheus, Grafana, GCP Pub/Sub, AWS SQS/SNS, Kafka, Vector Databases (pgvector, Qdrant), Memcache.

Applied AI

RAG architectures, Semantic Chunking, LLM Non-determinism, Vector Storage strategies, AI Agents, MCP servers.