Open Source Summit India 2026

Modern embedded audio devices increasingly run mainline Linux, but achieving deterministic low-latency multi-channel audio processing remains a challenge. While heterogeneous SoCs include DSP accelerators capable of handling signal processing workloads efficiently, integrating these accelerators into a standard Linux audio stack without proprietary middleware is non-trivial.
This talk presents a practical, upstream-friendly approach to building a zero-copy DSP offload framework using mainline Linux components such as remoteproc, rpmsg, and ALSA. Instead of relying on custom kernel patches or vendor-specific frameworks, the solution leverages DMA-backed shared memory and rpmsg-based signaling to enable efficient inter-processor communication between ARM application cores and a DSP.
A key focus of this session is eliminating redundant memory copies across kernel and user space boundaries. By designing a ping-pong buffer architecture with shared memory mapping and pointer-based synchronization, we achieved deterministic real-time streaming with significantly reduced CPU utilization and improved latency characteristics.

Modern Kubernetes platforms generate massive volumes of logs, events, metrics, and reconciliation signals, yet developers still struggle to answer a basic question: why did my deployment fail? At Comcast, operating large-scale internal Kubernetes platforms, failures often span platform abstractions, custom controllers, policies, and cluster runtime behavior—making manual diagnosis slow and unreliable.

This talk explores how AI can act as a diagnostic layer in Kubernetes platforms, focusing on failure explanation rather than detection. It presents architectural patterns for correlating platform intent with Kubernetes signals and translating complex control-plane behavior into clear, human-readable explanations and actionable guidance. Attendees will learn how AI can augment platform engineers, improve developer trust, and significantly reduce time-to-resolution without becoming another opaque system.

Robust guardrails are a foundational requirement for preparing AI agents for enterprise‑grade deployment. They ensure that agents consistently adhere to organizational standards, policies, and ethical expectations.

OpenTelemetry provides a vendor‑agnostic, standards‑based framework for observing and validating guardrail behavior at runtime. Its unified APIs and emerging generative AI semantic conventions enable organizations to safely instrument guardrail logic, eliminate silent failures, and capture valuable telemetry across decision points.

This session explores the role of guardrails across the entire AI stack, including data layers, infrastructure, orchestration components, and LLMs themselves. We demonstrate how open‑source frameworks such as NVIDIA NeMo Guardrails or OpenLIT can be used to implement production‑ready guardrail mechanisms. Finally, we show how key guardrail metrics such as pass/fail rates, top violation categories, and latency impact can be instrumented using OpenTelemetry and visualized through enterprise observability backends like Dynatrace or Grafana.

We all believe in open source - or we wouldn't be attending this conference. But although open source may be a "nice to have" property for software in general, this talk will try to convince you that security software really _must_ be open source. With nearly 30 years of open source and security experience, Mike will address some of the key ways in which the open source community does security - and also debunk a dangerous myth. We will ensure we have lots of times for questions - open source and security should both be two-way conversations!

Connected medical devices (IoMT) increasingly depend on Yocto‑based Embedded Linux platforms to perform critical, patient‑impacting functions. As cyberattacks on healthcare rise, securing these devices is now central to patient safety, regulatory compliance, and lifecycle quality. This session provides a practical, defense‑in‑depth blueprint for hardening IoMT devices—starting from secure boot and measured trust to OS‑level hardening, system integrity, secure OTA updates, sandboxing, runtime protection, and zero‑trust device‑to‑cloud communication.
We will map real‑world cyber trends to specific embedded mitigations and align them with global regulatory expectations for medical devices. The session also demonstrates how open‑source tools within the Yocto ecosystem can enforce reproducible security controls, generate SBOMs, and support automated vulnerability triage throughout the device lifecycle.
Attendees will gain engineering‑ready patterns to build secure, maintainable Linux‑based devices—applicable not only in healthcare, but also industrial IoT, automotive, and safety‑critical embedded domains.

Open source used to mean something simple: the code is open, the community builds it, and everyone benefits. That world is gone. Today, billion-dollar companies release model weights and call it "open source"

Projects launch with permissive licenses but lock their APIs behind paywalls. Foundations host projects where one vendor controls 95% of the commits. And a new generation of developers is entering open source through AI-generated pull requests they barely understand.

I've spent 7 consecutive Kubernetes release cycles on the release team, helped build and maintain Kargo - a OSS project for GitOps continuous delivery and worked as a CNCF Ambassador helping new contributors navigate this ecosystem

I've watched the definition of "open source" stretch, bend & sometimes break in real time.

This talk is about the real problems developers face today when they try to contribute to, depend on, or build careers around open source projects that don't play by the old rules. I'll share what I've learned about spotting "open-washing" evaluating project health beyond the GitHub star count, and building genuine community in an era where the incentives have fundamentally shifted.

Building modern applications in the cloud is exciting—but also overwhelming. You need CI/CD to ship fast, APIs to connect everything, and scalability to handle growth. But where do you start?

This session will break down the essentials of cloud-native development, covering CI/CD pipelines, API-driven architectures, and scalable deployments— while also showcasing how open-source cloud-native projects can accelerate your journey from beginner to pro.

You will learn:
1. CI/CD without confusion – Automate deployments with ArgoCD, Tekton & more.
2. APIs made simple – REST, GraphQL & event-driven APIs for cloud-native apps.
3. Scaling smartly – Kubernetes, Knative & serverless for effortless growth.
4. Open-source power-ups – CNCF projects that accelerate your development.
5. Best practices – The tools & workflows every cloud-native dev must know.

Join us to get a clear roadmap, hands-on tools, and the confidence to build, deploy, and scale like a pro!

Julian Gordon of LF Decentralized Trust will host a conversation with a cross-section of Indian organizations that are driving digital public infrastructure implementations, both domestically and internationally, including Networks for Humanity and Digi Yatra Foundation. India is leading the way in transforming digital ecosystems in government, finance, identity, and more. This session will be an opportunity to hear from those developing and deploying critical open source decentralized technologies to modernize core systems, ushering in more efficiencies, enhanced transparency, stronger security, and increased privacy.

Key topics will include:

-Real-world cases in banking and payment networks, government data infrastructure, decentralized identities, and energy transition
-The role of open development and governance in fostering the collaboration, innovation and trust required for these deployments
-Challenges including interoperability, security, regulatory compliance and skill gaps
-Practical insights for building resilient, collaborative, future-proofed open source networks

Murphy's Law states, "Anything that can go wrong will go wrong." For an Open-Source project, this can be a daunting task. Imagine you are in between a major architectural change, trying to decouple a monolith and your community itself starts to decouple. This is what happened recently in our decade old community and we would like to share the raw behind-the-scenes of our journey, hoping it will be helpful for others who're going through the same.

Our story will portray, how we lost our veteran leaders and architects departing the community, placing an unprecedented load on the remaining maintainers to bridge a massive gap in expertise, institutional knowledge and maintain stability while executing a high-stakes architectural migration. We will also touch bases on how the GenAI can work as a double-edge sword when you are in short of contributors.

Key Takeaways:
1. Practical strategies for project survival when veteran institutional memory departs mid-migration.
2. Why Next.js was a strategic necessity and not just a nice-to-have upgrade.
3. Navigating the influx of high-volume, variable-quality contributions and managing the burnout.
4. Lessons on building a new leadership.

This session presents the challenges which Telco service providers have in moving towards Autonomous Networks and adopting GenAI, AgenticAI - especially for OSS and Network Operations. This presentation highlights how some of the challenges can be solved using LFN projects and benefits to Telco Network Service Providers

Learn how to run real AI models fully on-device using Qualcomm’s AI Engine with LiteRT and ExecuTorch. This talk covers architecture, deployment pipelines, performance tuning on NPUs, and real demos showing how to achieve fast, private, low-latency inference on modern AI PCs and mobile devices.