Sat Agrawal
Senior Principal Software Engineer @ Discover Financial Services
Jacksonville, Florida, United States
Actions
Sat Agrawal has an unvarnished commitment to delivering quality results. An avid learner, Sat holds a bachelor’s in computer science from Penn State and remains actively engaged in a wide range of learning resources. He is a Certified Chaos Engineering professional and practitioner, with additional certifications that include CKA, CKAD, and TOGAF. Sat's current focus in service mesh technologies exemplifies his commitment to stay at the forefront of cloud-native technologies.
At Discover Financial Services, Sat combines his diverse knowledge of concepts, practices, and processes with a strong advocacy for automation and efficiency. Tapping native strengths in analytical and communication skills, he enters mentoring and partnering relationships with business users, architects, and stakeholders across the enterprise to help product teams streamline their end-to-end product delivery cycles. A recognized top-level contributor himself, Sat influences engineers to embrace the agile practices that transform the way they work, leading ultimately to the delivery of dynamic, impactful solutions that align with organizational goals.
Sat personalizes his aptitude for leadership through his attendance and speaking at conferences and local meetups and by conducting hands-on proofs-of-concept. Participating in LinkedIn and other industry-leading blogs, he engages in a variety of topics that include DevOps, DevSecOps, and SREs.
Area of Expertise
Topics
Unleashing the Power of Observability for Building Resilient and Efficient Systems
Cloud native observability is not just about monitoring metrics and logs; it encompasses a holistic approach to understanding the intricate interactions and dependencies within distributed systems.
For financial services organisations, the ability to gain deep insights into the behaviour of interconnected services is crucial for maintaining system reliability, detecting anomalies, and optimizing performance. However, navigating the complexities of observability can be daunting, with a plethora of tools, techniques, and best practices to consider.
Prometheus and Grafana have emerged as a powerful open source combination for cloud native observability, enabling efficient monitoring, data collection, and visualization.
This talk will guide attendees through the integration of Prometheus and Grafana, and showcase real-world examples for advanced observability techniques, such as anomaly detection, auto-remediation, and predictive analytics.
The art of getting to less - Embracing the agile principle of simplicity
Come hear about applying one of the trickiest Agile principles: "Simplicity--the art of maximizing the amount of work not done--is essential"! This agile principle is frequently overlooked when organizations and teams create or improve products. Why do humans love to add in order to improve? This gets in the way of faster, smoother and better delivery. Hear about the psychology behind why we do this. Learn how organizations like Discover use dojos and academies to helps teams improve.
Take Control of Your Infrastructure: Code It for Success
Ready to take charge of your project's infrastructure? This session introduces you to Infrastructure as Code (IaC), the key to efficient and understandable infrastructure management for any software initiative. Learn how IaC replaces manual processes with code, ensuring consistency and repeatability across all your environments, in any kind of project from startups to large enterprises. We’ll break down the core concepts and demonstrate how IaC puts you in control, streamlining operations and improving team collaboration, no matter the type of project you’re working on. For maximum efficiency and automation, imagine applying your IaC definitions automatically whenever you need to update your infrastructure - that's the power of combining IaC with Continuous Delivery.
Pipeline Power-Up: GitOps for Scalable Ecosystems
GitOps simplifies CD, yet scaling pipelines in complex ecosystems demands more. This session explores advanced GitOps patterns for large deployments, multi-tenancy, and robust security. We'll cover:
- Scalable pipeline architectures: Optimized for GitOps efficiency and handling growth.
- Ecosystem-focused GitOps: Enabling easier contribution, collaboration & sharing best practices across teams & projects in the CD ecosystem.
- Security & Compliance: Integrating seamlessly into GitOps for broader ecosystem trust and security standards.
Our Reliability Revolution - How we built a culture of SLO driven decision making
Tools don't solve reliability problems, people do! Ensuring we have effective processes and a shared understanding of our reliability goals are more crucial to creating a holistic assessment of your products reliability than the tools used to observe it. Despite a wealth of available monitoring tools, leaders can lack a common framework for defining reliability, balancing investments, and setting expectations. By applying Site Reliability Engineering (SRE) principles like Service Level Objectives (SLOs) and error budgets to create a structures, data-driven approach, Immersive learning experiences and embedded coaching, product leaders and engineers gained practical skills to help set baselines, use error budgets for prioritization, and automate monitoring with InnerSource practices. Join us to explore how we standardize reliability expectations within large organizations and help transform reliability from an abstract goal into an organizational practice.
Optimizing AI/ML Workloads on Kubernetes: Cutting Costs Without Compromising Scale
Unleash the power of distributed AI/ML training on Kubernetes without breaking the bank. Discover open-source gems like Kubernetes cluster autoscalers, DASK, and Volcano that unlock intelligent scheduling, autoscaling, and resource optimization across your cluster. Explore real-world case studies on maximizing GPU utilization, right-sizing resources, and leveraging spot instances to their full potential.
Cut your compute costs by up to 60% while maintaining peak performance. Gain valuable insights into evaluating and adopting cost-effective distributed training frameworks such as Horovod, TensorFlow Distributed, and PyTorch Lightning, tailored for Kubernetes environments. Leave with actionable strategies to optimize your AI/ML pipelines for both scalability and cost-efficiency on any cloud platform.
Designing Resilient Cloud Native Architectures For Financial Services Applications
Service mesh presents distributed systems with a comprehensive solution for managing intricacies in network traffic, security, observability, and resilience. Within the dynamic, yet complex landscape of financial services, many organizations utilize service mesh to establish and maintain secure communication for significant array of microservices. Adopting such security measure ultimately results in strengthening IAM, enabling encryption, enforcing security policies, and facilitating compliance monitoring.
Many financial services organizations, however, face challenges in adopting service mesh for their cloud native modernization approach.
This talk aims to explore the real-world use cases and lessons learned in the realm of service mesh, with special emphasis on its integration with the Kubernetes ecosystem, as well as configuring and managing service mesh policies for financial services applications, empowering to build scalable, resilient, and efficient microservice architectures.
Becoming a Stronger Ally : Handling Difficult Situations and Building Inclusive Communities
Dojos help provide a safe space which encourages active listening to the experiences of marginalized groups, educate about their perspectives, build empathy towards those different from you. This interactive session provides an arena for participants to practice allyship skills and consider more diverse perspectives. Through facilitated discussions and a scenario-based exercise (replacing the "Dojo" context), participants to learn how to actively listen to the experiences of marginalized groups, understand diverse perspectives, and build empathy. This session aims to equip you with practical strategies for navigating complicated situations and fostering a more inclusive open-source community
Serverless Unleashed: Powering the Future of Cloud-Native Innovation
Dive into the exciting world of serverless computing and discover how it's revolutionizing application development. This dynamic session will demystify serverless architecture, showcasing its potential to streamline operations, boost scalability, and drive innovation. We'll explore real-world use cases, from AI-powered chatbots to event-driven microservices, demonstrating how serverless is reshaping industries. Get hands-on with popular open-source serverless frameworks and learn best practices for building, deploying, and managing serverless applications. Whether you're a curious beginner or a seasoned developer looking to expand your toolkit, this session will equip you with the knowledge to harness the power of serverless and stay ahead in the rapidly evolving cloud-native landscape.
Optimizing Microservices: Creating Efficient OCI-Compliant Containers with Podman
Learn how to create and optimize OCI-compliant container images for microservices using Podman and other open-source tools. This session will highlight the benefits of OCI compliance for enhanced container portability and runtime flexibility. We will cover best practices for building efficient container images, including the use of minimal base images and strategic layer structuring to improve startup times. Through a practical books management microservice example, we will demonstrate benchmarking techniques to compare the performance of optimized versus non-optimized images. Join us to refine your containerization approach, ensuring your microservices are lightweight, efficient, and ready for any runtime environment.
Which technical practices naturally foster a culture of trust and accountability?
Learn about how technical transformation and dedication to excellent DevOps practices precede any organizational revolution when looking to change with purpose. Although cultural pillars of high performing teams like trust and accountability are widely sought after, certain DevOps practices help lead to these tendencies more naturally, enabling leadership to quickly see the fruits of their support and engineers to work autonomously, avoid toil and be able to rely on each other to achieve the best outcomes.
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