Anupam Phoghat
Principal
Sydney, Australia
Actions
I am a seasoned IT professional with years of experience in the field. I have extensive experience with Google Cloud Platform (GCP), and currently hold many professional level Google Cloud and other industry certifications. I am also certified in Kubernetes, and have a deep understanding of the core concepts of Kubernetes.
I am passionate about sharing my knowledge with others by actively participating in GCP community. This also helps me in my own learning and growth in the field.
Area of Expertise
Topics
Unleashing Bounded Contexts: Serverless Microservices with DDD on Google Cloud Run and GKE Autopilot
In the ever-evolving world of microservices, Domain-Driven Design (DDD) offers a powerful framework for taming complexity and aligning software with business realities. This talk takes you on a journey into the core of DDD, where we'll explore how to leverage Google Cloud's serverless powerhouses—Cloud Run and GKE Autopilot—to build and deploy highly scalable, resilient, and cost-effective microservices.
Under the Hood of Agentic AI: Practical Monitoring and Evaluation in Vertex AI
Building an AI agent is only the first step; keeping it reliable, accurate, and cost-effective in production is the real engineering challenge. This session dives deep into the operational realities of Agentic AI. We will step away from basic prompt engineering and focus entirely on the monitoring stack. Using Vertex AI, we will implement practical techniques for evaluating agent performance, logging reasoning paths, and monitoring tool execution in real-time, ensuring your agents do exactly what they were designed to do without unexpected deviations.
The Secret Handshake: Vault and Service Mesh Security
Microservices have revolutionised application development, but they've also introduced new security challenges. The service mesh, a powerful tool for managing and securing communication between these services. But how do we ensure that only authorised services can access sensitive information? That's where HashiCorp Vault comes in.
In this talk, we'll explore the "secret handshake" between Vault and service mesh technologies like Istio and Linkerd. We'll demonstrate how to leverage Vault's robust secrets management capabilities to secure your service-to-service communication.
The Telemetry Trap: Building Meaningful Open-Source Observability for Multi-Agent Systems
Monitoring traditional microservices is a solved problem. We know how to track CPU, memory, and latency. But how do you monitor the "reasoning" of an AI agent? If you can't measure the cost and logic of your GenAI workflows, you are flying blind in production.
This session explores how to adapt the open-source monitoring tools you already use (like Prometheus and OpenTelemetry) to track the health, infrastructure costs, and hallucination rates of agentic workflows. We will treat LLM calls just like any other microservice dependency, demonstrating how to instrument an AI pipeline, map token usage directly to infrastructure spend, and build dashboards that expose both system health and AI accuracy in real-time. Stop guessing what your agents are doing and start monitoring them like proper engineering assets.
The LLMOps Telemetry Trap: Instrumenting Multi-Agent Systems with OpenTelemetry and Prometheus
Monitoring traditional microservices is a solved problem. We know how to track CPU, memory, and latency. But how do you monitor the "reasoning" and execution loops of an AI agent? If you can't measure the token cost, logic paths, and hallucination rates of your GenAI workflows, you are flying blind in production.
This session explores how to adapt the open-source monitoring tools you already use—specifically Prometheus and OpenTelemetry—to track the health and infrastructure costs of agentic workflows. We will treat LLM calls just like any other distributed microservice dependency. I will demonstrate how to instrument an AI pipeline, map token usage directly to infrastructure spend, and build Grafana dashboards that expose both system health and model performance in real-time. Stop guessing what your agents are doing and start monitoring them like proper engineering assets.
The Intelligent Endpoint: From Data Warehouse to GenAI-Powered API
Go beyond CRUD. This session provides a strategic roadmap for transforming dormant data assets into active, intelligent API products. Learn the architectural patterns and security considerations for taking your data from a cloud warehouse, training a GenAI model, and deploying it as a secure, monetisable, and intelligent API endpoint.
The AI Governor: A Playbook for Securely Scaling AI APIs
This talk is for leaders in larger organisations who are trying to enable their teams with AI without creating chaos. The core idea is that you can't have a hundred teams connecting to a hundred different AI services with no oversight. You need a strategy, and APIs are the key.
Centralise & Conquer: Create a centralised, internal "AI Gateway" API. This single point of entry allows the company to manage costs, enforce policies, and maintain security.
Security & Governance: How this central API layer is the perfect place to implement security controls, audit trails, data loss prevention (DLP), and enforce ethical AI principles.
Empowerment: Not a bottleneck, but a platform that empowers developers to safely and securely leverage AI capabilities without needing to be security or AI experts themselves.
Taming the chaos: Sentinel with Terraform workflows in Multi-Cloud environments
Tired of infrastructure drift and security nightmares? In this session, we'll dive deep into a common challenge faced by many organizations: ensuring consistent security and compliance across their cloud infrastructure.
We'll explore how Sentinel, HashiCorp's policy-as-code framework, can be integrated with Terraform to proactively enforce policies and prevent misconfigurations before they reach production. Through real-world examples and live demos, you'll learn how we managed multi tenants deployment by
- Define and implement custom Sentinel policies
- Integrate Sentinel with Terraform workflows
- Manage policy exceptions and overrides
- Leverage Sentinel's testing framework
I'll also share lessons learned and best practices from my own experience using Sentinel and Terraform to manage infrastructure at scale. You'll leave this session with practical techniques and actionable insights to help you tame the chaos and build a more secure and compliant cloud environment.
Supercharge Your Developer Velocity with Automated Deployment and Management on GKE Autopilot
In this session, titled “Supercharge Your Developer Velocity with Automated Deployment and Management on GKE Autopilot,” we’ll dive into how to significantly enhance your development workflows by leveraging the power of Google Kubernetes Engine (GKE) Autopilot in tandem with Helm charts. This technical talk is designed for DevOps engineers, cloud architects, and developers who are eager to streamline their application deployment processes while maintaining robust security and best practices.
We’ll explore how GKE Autopilot, with its fully managed Kubernetes experience, abstracts away the operational complexities, allowing teams to focus on accelerating development cycles. Through the use of Helm charts, you’ll learn how to automate deployments, manage upgrades seamlessly, and integrate CI/CD pipelines for continuous delivery.
My Server's Epic Journey: From a Simple App to a Scalable World
Ever wanted to spin up a game server for your friends but got lost choosing between serverless, Kubernetes, and all the other options? Join us on a fast-paced, epic journey to deploy a Minecraft server on Google Cloud. We'll start with the simplicity of Cloud Run, discover why GKE is the hero for stateful workloads, and even peek at the magic of adding AI with Vertex AI. You'll leave with a clear mental map for choosing the most efficient and powerful tool for your next project.
Redstone vs. Diamond: Balancing Cost, Complexity & Power on Google Cloud
Using a story-based approach, we'll experience key lessons via failure and success while choosing between cloud run, GKE and VertexAI to deploy agents.
Cloud Run is Redstone: Common, simple to use, and incredibly efficient for specific, automated tasks (like a simple webhook).
GKE is a Diamond Pickaxe: It requires more "investment" to craft (i.e., more learning and setup), but it's the powerful, durable tool you need for the big, tough jobs (like running your main game server).
Vertex AI is an Enchanting Table: A specialised, high-powered tool that you bring in to add magical abilities (AI features) to your existing tools. It's not for mining, but for making your mining better!
Integrating DevSecOps with HashiCorp: A practical guide to securing your infrastructure
This talk will explore the practical aspects of implementing DevSecOps practices using HashiCorp tools such as Terraform, Vault, and Consul. Attendees will learn how to automate security within their infrastructure as code workflows, ensuring that security is a foundational component rather than an afterthought. The session will cover real-world scenarios, including automating secrets management with Vault, enforcing compliance through Terraform, and utilizing Consul for secure service networking.
Instrumenting the Cloud: Eliminating Blind Spots with OpenTelemetry
Troubleshooting distributed, cloud-native systems shouldn't feel like a guessing game. This code-heavy, practical session focuses purely on the mechanics of capturing high-quality telemetry data. We will instrument a microservices architecture to generate custom logs, metrics, and distributed traces using OpenTelemetry. By the end of the session, attendees will understand how to properly instrument their applications and route that telemetry to centralised observability backends to instantly pinpoint bottlenecks and silent failures.
Instrumenting the Cloud: Eliminating Blind Spots with OpenTelemetry
Troubleshooting distributed, cloud-native systems shouldn't feel like a guessing game. This code-heavy, practical session focuses purely on the mechanics of capturing high-quality telemetry data. We will instrument a microservices architecture to generate custom logs, metrics, and distributed traces using OpenTelemetry. By the end of the session, attendees will understand how to properly instrument their applications and route that telemetry to centralised observability backends to instantly pinpoint bottlenecks and silent failures.
GenAI to Life - Build Your Own AI-Powered Learning Assistant
In this hands-on workshop, you'll build an AI-powered application that generates personalised learning materials using Google's Gemini model. By specifying a topic, learning style (visual, auditory, kinesthetic), and desired output (summary, quiz, examples), participants will create dynamic educational content tailored to individual needs.
This session will guide you through prompt engineering techniques, AI content adaptation, and deployment on Google Cloud, showcasing how generative AI can transform learning experiences.
From Ubiquitous Language to Serverless APIs: Building Domain-Driven APIs with Google Cloud Endpoints
In the world of Domain-Driven Design, a shared Ubiquitous Language is key to bridging the gap between domain experts and developers. But how do we translate this language into clean, scalable, and maintainable APIs? This talk will delve into the art of crafting Domain-Driven APIs using Google Cloud Endpoints. We'll explore:
The Essence of a Ubiquitous Language: We'll start by understanding the importance of a shared language in DDD and how it shapes our APIs.
Modeling APIs with DDD: We'll dive into practical techniques for mapping DDD concepts like Aggregates, Entities, and Value Objects to API resources and endpoints.
Leveraging Google Cloud Endpoints: We'll harness the power of Google Cloud Endpoints to streamline API development, deployment, and management. We'll see how Endpoints enables features like API versioning, authentication, and monitoring, making our APIs robust and secure.
Serverless Scalability: We'll explore how to leverage Google Cloud Functions to create serverless API backends, ensuring our APIs can effortlessly handle varying loads while keeping costs in check.
Demystifying Modernisation: True Potential of Cloud Technology
In the fast-paced digital landscape, businesses are increasingly looking to leverage the true potential of cloud technology through digital modernization. This speech will focus on Application Modernisation on the Google Cloud Platform (GCP) using cutting edge tools from Hashicorp and how it is closely tied to Infrastructure Modernisation and Data Modernisation. The speech will explore the three distinct categories of modernization: Infrastructure Modernisation, Data Modernisation, and Application Modernisation.
Collaboration with Gemini on GKE
Learn how to interact with gemini with proper prompts using agents running on GKE
Beyond the Batch: Real-Time Schema Mapping with Gemini, Cloud Run, and dbt
Modern data engineering requires agility that traditional batch processing struggles to provide. This hands-on talk tackles the complex challenge of wrangling unpredictable, unstructured data streams into actionable formats. Through a live architectural walkthrough, we will demonstrate how to combine Gemini's reasoning capabilities with Cloud Run to execute dynamic, real-time schema mapping. Finally, we will integrate dbt as an external orchestrator to handle the subsequent data transformations, providing a complete, production-ready blueprint for modern data pipelines.
Build Your Own AI-Powered Learning Assistant with Gemini
In this hands-on workshop, you'll build an AI-powered application that generates personalised learning materials using Google's Gemini Pro model. By specifying a topic, learning style (visual, auditory, or kinesthetic), and desired output (summary, quiz, or examples), participants will create dynamic educational content tailored to their individual needs.
This session will guide you through prompt engineering techniques, AI content adaptation, and deployment on Google Cloud, showcasing how generative AI can transform learning experiences.
APIs as the New Frontier: Monetising and Securing Your AI
This is a top-level business and technology strategy talk. Instead of just showing how to build an AI API, we'll discuss why and what a company needs to consider.
The Opportunity: Looks at AI models not as internal projects, but as strategic assets that can be productised and monetised through APIs.
The Architecture: High-level patterns for exposing AI—when to use a pre-trained model API (like Gemini) vs. when to fine-tune and host your own.
The Threat: The unique security challenges of AI APIs: prompt injection, data poisoning, model inversion attacks, and the massive cost implications of abuse. It's not just about a standard 403 Forbidden error anymore.
Architecting the AI Boardroom: A Multi-Agent System for a Startup Founder
How do we move from simple chatbots to complex AI systems that solve genuine industry problems? Drawing on the concept of an "AI C-Suite," this session explores the architecture and deployment of a multi-agent system designed specifically to tackle operational and educational challenges in K-12 learning environments. We will map out and build a workflow where specialized, autonomous agents—such as a personalized learning assistant, a scheduling optimizer, and an administrative analyst—collaborate to streamline operations and enhance educational outcomes.
Agentic SRE: Not Alerting; Autonomous Remediation on GKE
Modern infrastructure generates more alerts than human engineers can ever realistically handle. We are stuck in a loop of getting paged, checking dashboards, and executing runbooks. It’s time to move past static alerting and build systems that can actually fix themselves.
This session bridges the gap between traditional infrastructure-as-code and Generative AI. I will demonstrate how to build an "Agentic SRE" workflow that doesn't just page you—it investigates. We will walk through a practical, production-ready architecture using Cloud Run to catch Alertmanager webhooks, passing the payload to Gemini (via Vertex AI) to query open-source telemetry for context, and finally proposing (or executing) safe infrastructure-as-code fixes via Terraform and kubectl. This isn't sci-fi; it's just smart orchestration. You will leave with a clear roadmap for implementing AI-driven remediation in your own Kubernetes clusters.
Agentic SRE: Moving from Alerting to Autonomous Remediation in Day-2 Operations
Modern cloud-native infrastructure generates more alerts than human engineers can ever realistically handle. We are stuck in a loop of getting paged, checking Grafana, and executing runbooks. It’s time to move past static alerting and build systems that can actually fix themselves.
This session bridges the gap between traditional infrastructure-as-code and Generative AI for Day-2 operations. I will demonstrate how to build an "Agentic SRE" workflow that doesn't just page you—it investigates. We will walk through a practical, production-ready event-driven architecture that catches Prometheus webhooks, passes the payload to an LLM to query open-source telemetry for context, and finally proposes safe infrastructure-as-code fixes via Terraform and kubectl. This isn't sci-fi; it's just smart orchestration. You will leave with a clear roadmap for implementing AI-driven incident response in your own high-scale Kubernetes environments.
300 Seconds of Chaos to Order: The Self-Curation Engine Micro-Demo
Data schemas are messy, unpredictable, and constantly breaking downstream pipelines. In this rapid-fire micro-demo, we are going to fix that on the fly.
With zero slides and pure terminal action, I will demonstrate the "Self-Curation Engine"—a highly actionable architectural pattern for dealing with chaotic JSON payloads. We will use Cloud Run to catch unstructured data and leverage Gemini to map the schema in real-time before it hits the data warehouse. It’s 300 seconds of pure engineering value, showing exactly how to use GenAI for robust data engineering.
Advanced PromQL Hacks Every Cloud-Native SRE Needs for Day-2 Ops
When your multi-cluster Kubernetes environment starts acting up, you don't have time to read the manual. You need answers, immediately.
This lightning talk delivers 10 minutes of pure, unadulterated engineering value for Day-2 operations. I will walk you through three highly specific, underutilized PromQL queries and open-source observability configurations that instantly cut through the noise and make debugging production clusters easier. These are the exact copy-pasteable techniques we use in the trenches to identify deep network bottlenecks, hidden CPU throttling, and cascading pod failures across hybrid systems.
From Manual Assembly to High-Speed Automation: Ensuring Secure Deployments
Platform engineers are constantly challenged with streamlining application deployments while maintaining robust security. This session dives deep into leveraging automation and zero-trust principles to achieve both goals.
We'll explore how automated deployment of CI/CD and security tooling in a kubernetes DevOps cluster running on Google Cloud Platform reduced the time taken for the whole application deployment process 70%. We'll also look at the concept of zero trust and its application in securing the deployment pipeline. We'll demonstrate how to leverage open-source tools like ArgoCD and Kustomize to automate deployments with least privilege access across the entire CI/CD chain. Additionally, the talk will cover integrating security scanning tools within the pipeline for early detection and mitigation of vulnerabilities.
Unleashing Scalability and Flexibility: A Deep Dive into Google Cloud Platform's Serverless Service
Serverless is hot, everyone knows it. But when it comes to taking the plunge, the "how" can seem daunting. Join me, a fellow developer from the ANZ community, as I share my real-world journey building, deploying, and scaling applications with one of Google Cloud Platform's (GCP) most dynamic offerings - Cloud Run. In this technical talk, we will delve into the intricacies of Cloud Run, uncovering its architecture, key features, and real-world applications.
This talk isn't just a technical deep dive (though we'll get there!); it's a tale of trenches, demons, and ultimately, triumph. You'll hear:
The challenges: What pain points usually lead to explore serverless?
The solutions: How does Cloud Run and other managed services help overcome those challenges?
The lessons learned: What valuable insights did I gain along the way?
Demystifying Modernisation: True Potential of Cloud Technology
In the fast-paced digital landscape, businesses are increasingly looking to leverage the true potential of cloud technology through digital modernization. This speech will focus on Application Modernisation on the Google Cloud Platform (GCP) using cutting edge tools from Hashicorp and how it is closely tied to Infrastructure Modernisation and Data Modernisation. The speech will explore the three distinct categories of modernization: Infrastructure Modernisation, Data Modernisation, and Application Modernisation.
KSUG.AI Australia #67 - Melbourne Meetup @AWS - 2 Sep 2026 User group Sessionize Event Upcoming
DevFest Taipei 2025 Sessionize Event
Devfest Brisbane 2025 Sessionize Event
GDG Melbourne Devfest 2025 Sessionize Event
90DaysOfDevOps - 2024 Community Edition Sessionize Event
HashiTalks: India Sessionize Event
Anupam Phoghat
Principal
Sydney, Australia
Actions
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top