DeveloperWeek 2025

Moshe Twitto, Pliops, Founder & CTO

Organizations are increasingly concerned about the lack of power budgets in data centers, particularly as AI infrastructure and emerging AI applications lead to higher energy footprints and strain cooling systems. As they scale their AI operations and add GPU compute tiers, the escalating power and cooling demands, coupled with significant capital investments in GPUs, are eroding margins. A monumental challenge looms as data centers struggle to secure essential power, creating significant pressure for companies striving to expand their AI capabilities.

In today's LLM inferencing computing, GPU prefill operations are heavily compute-bound and critically determine the batch size. While prefill can fully utilize GPU resources, increasing the batch size beyond a certain point only increases the Time to First Token (TTFT) without improving prefill rate. On the other hand, GPU decode operations are HBM bandwidth-bound and mainly influenced by model and KV cache sizes, benefiting significantly from larger batch sizes through higher HBM bandwidth efficiency. Pliops' solution improves prefill time, allowing for larger batch sizes without violating user SLA for prefill operations. This enhancement directly affects decode performance as well, as it gains greatly from the increased batch size. As a result, by improving prefill time, the system achieves nearly proportional improvements in end-to-end throughput.

Shambhavi Srivastava, Appen, AI Solutions Architect

Data annotation involves assigning relevant information to raw data to enhance machine learning (ML) model performance. While this process is crucial, it can be time-consuming and expensive. The emergence of Large Language Models (LLMs) offers a unique opportunity to automate data annotation. However, the complexity of data annotation, stemming from unclear task instructions and subjective human judgment on equivocal data points, presents challenges that are not immediately apparent.

In this session, Chris Stephens, Field CTO and Head of AI Solutions at Appen will provide a overview of an experiment that the company recently conducted to test the tradeoff between quality and cost of training ML models via LLMs vs human input. Their goal was to differentiate between utterances that could be confidently annotated by LLMs, and those that required human intervention. This differentiation was crucial to ensure a diverse range of opinions or to prevent incorrect responses from overly general models. Chris will walk audience members through the dataset used as well as methodology for the experiment, as well as the company’s research findings.

 

Parth Shukla, Cequence Security, Security Analyst
Khyati Ganatra, Cequence Security,  Manager, Applied ML 

In the presentation titled "AI Frontiers: Shielding Digital Gateways from Bot Invasions," we delve into the forefront of cyber defense against bot-driven threats that exploit API vulnerabilities. This comprehensive study explores how advanced AI and ML models are being harnessed to fortify digital defenses, offering a detailed analysis of API communication patterns and the evolving landscape of bot attacks. Through a series of real-world case studies, we illuminate the mechanisms of sophisticated bot strategies—ranging from data breaches and account takeovers to shopping bots that deplete inventories. The narrative progresses to unveil how AI/ML technologies serve as the cornerstone of innovative defense mechanisms. We dissect the architecture of AI-driven systems tailored to detect and counteract anomalous behaviors indicative of bot activities, leveraging vast datasets to train ML models that adeptly differentiate between legitimate user interactions and malicious bot intrusions. The discussion further navigates through the technical and operational nuances of implementing AI/ML defenses, emphasizing predictive analytics for preemptive action, machine learning for dynamic threat adaptation, and the overarching impact of such technologies in securing digital ecosystems against the insidious threats posed by automated attacks. This presentation not only highlights the challenges but also showcases the resilience and adaptability of AI/ML solutions in the ever-evolving battle against digital villains.

Animesh Mishra, Diffblue, Senior Solutions Engineer

Developer toil is real — days, even weeks, are spent conducting manual, repetitive, time-gobbling tasks like unit testing. And while GenAI coding assistants can support developers, they repeatedly fall short in performing complex TestQA tasks without human oversight.

The problem is compounded by the shortage of experienced testers, leaving developers drowning in QA testing rather than writing and innovating application code. Reports show 75% of developers’ time is spent on tasks other than code generation, resulting in 57% of developers admitting to skipping unit tests entirely.

For companies to succeed, code must be fool proof. So how can developers ensure testing is comprehensive and done with due diligence — and at scale?

The answer lies in reinforcement learning-based autonomous AI, which can eliminate up to 95% of the time developers typically spend on test writing. In this session, you will learn how to eliminate developer toil by automating manual and tedious test operations at scale (both locally and in CI), how the right tools can autonomously write and maintain unit tests for entire Java code bases, and how to ensure your unit tests for complex code are comprehensive.

This talk is designed for dev leaders looking to push the boundaries of what’s possible in Test/QA with emerging AI technology.

Oz Katz, Treeverse, CTO & Co-Founder

Langchain has become one of the most popular frameworks for anyone building custom, generative AI-driven apps powered by LLMs, that leverage RAG for the most enhanced results. But like all data products, these applications are really only as good as the org data fed into them.

In this live session you’ll learn how to build a reproducible AI app pipeline with Langchain & lakeFS, including how to build a RAG chatbot, while iteratively tuning it for best results leveraging Delta Lake’s temporal versions, you’ll come away with improved methods for data reproducibility for custom AI apps.

Jon Peck, GitHub, Technical Advocate & Software Developer

Software development is a team sport, with many different roles, where eveyone can win. But success isn't guaranteed; it depends on specific practices, policies, and tools which enable minimally-siloed, AI-accelerated collaboration across all parts of the DevOps process, from PM to development to CI/CD and security.

Discover the patterns and tools which lead to success, methods for changing the status quo, and perhaps a few horror stories. We'll touch on innersourcing, cloud development, AI, automation, governance, security, scaling and more -- with actionable learnings for everyone from small maintainer communities to F500 Enterprises.

Jothimani kanthan Ganapathi, Intuit, Principal Engineer
Anish Dhason, Intuit, Group Product Manager

Intuit, a global leader in financial software, is revolutionizing the industry by integrating Generative AI (GenAI) across its entire product portfolio, including TurboTax, QuickBooks, Mailchimp, and CreditKarma. By embedding GenAI into critical aspects of the development life cycle—such as code generation, review, and test creation—Intuit is driving innovation, enhancing productivity, and delivering smarter, more efficient experiences for both developers and customers alike.

This presentation explores how Intuit is revolutionizing the software development lifecycle by leveraging Generative AI (GenAI) to accelerate application development, enhance developer productivity, and streamline complex coding tasks. A key focus will be on how Intuit has implemented secure and scalable methods to integrate Intuit-specific knowledge into the context of GenAI-assisted developer tools. Through real-world case studies, we will demonstrate how AI-powered developer assistants are automating code generation for Intuit-specific APIs, commonly used libraries, and unit test creation, leading to a 25% reduction in task completion times with AI-driven tools.

Join us to discover how Intuit is pushing the boundaries of what's possible in modern software development, driving both innovation and efficiency using GenAI

Dev Roundtables / Product Roundtables / AI Roundtables give attendees the opportunity to meet in small groups with DeveloperWeek + ProductWorld + AI DevWorld speakers and industry leaders to discuss real-world topics relevant to you. These open roundtables are working sessions for in-depth industry conversations.

Alison Cossette, Neo4j, Data Science Strategist, Advocate, Educator

As generative AI continues to push boundaries, creating novel content by learning from massive datasets, we are faced with complex issues around intellectual property, privacy, and the ethical use of data. Current systems of copyright, fair use, and data protection lack the scope to fully address the unique challenges posed by AI pattern recognition and generation.

This pivotal talk introduces the pioneering concept of "Pattern Rights" - a holistic ethical framework to inform the development and deployment of generative AI technologies. Pattern Rights serves as an umbrella construct, encompassing principles of copyright, fair use, training data transparency, privacy, data ownership, and accountability.

We will explore how Pattern Rights can ensure appropriate attribution and compensation when AI models learn from copyrighted works or personal data. It establishes guidelines around consent, anonymization, and ethical data sourcing practices.

As the AI industry is rapidly evolving, we urgently need governance to foster innovation while upholding rights and safeguarding against misuse. Pattern Rights provides a roadmap to navigate this uncharted territory responsibly and equitably.

Stephen Chin, Neo4j, VP of Developer Relations

The human brain is divided into two hemispheres, each of which specializes in different types of tasks. Your right brain excels at imagination, intuition, art, rhythm, feelings, and dreams. Meanwhile, the left brain is more analytical and focused on logic, sequencing, linear thinking, and mathematics. Like a human brain, generative AI excels at certain tasks, especially those related to right-brain creative activities. However, the absence of relevant and up-to-date information produces confusing hallucinations that are not grounded in reality. 
This presentation will explore how knowledge graphs built with factual and updated information can complement LLMs via retrieval augment generation (RAG) to produce more reliable results.

Dima Spivak, IBM, Vice President of StreamSets Development & Product Management

Modern enterprises operate in a fast-paced digital landscape where acting on the latest data is essential to staying competitive. IBM StreamSets is a powerful, developer-friendly platform designed to address the challenges of streaming data integration across hybrid and multi-cloud environments. In this session, we’ll explore how StreamSets simplifies the ingestion, transformation, and delivery of structured, unstructured, and semi-structured data using an intuitive no-code/low-code interface.

Discover how adaptive pipelines handle schema drift, optimize data quality, and ensure low-latency data movement for mission-critical workloads. We’ll also showcase the platform’s flexible deployment options and seamless integration with popular technologies like Apache Kafka and Pulsar, enabling developers to rapidly operationalize streaming data pipelines with minimal overhead.

Dan Edwards, Crowe, Managing Director

In an era where artificial intelligence (AI) and machine learning are rapidly evolving, the accounting profession stands at the precipice of significant transformation. This session aims to demystify the impact of AI technologies, such as Microsoft Copilot, on the future of accounting.

Join us for an insightful exploration into the capabilities of AI in the accounting space, where we will delve into the current state of AI tools like Microsoft Copilot, their potential to automate complex accounting tasks, and the implications for accounting professionals.

This session is designed for accounting professionals at all levels who are seeking to understand the impact of AI on their careers and how to strategically position themselves for success in a rapidly changing technological landscape. Whether you're a seasoned CPA or a recent graduate, this session will provide valuable insights into the future of accounting.

Utkarsh Contractor, Aisera, VP of AI

Enterprises are on the cusp of a transformative shift. They are moving beyond task-specific AI to embrace agentic systems capable of autonomous decision-making and actions through probabilistic reasoning. This evolution marks a new era of autonomy, where AI not only augments human capabilities but also drives business outcomes by:
- Optimizing Operations: Self-regulating systems enhance efficiency, ensure compliance, and proactively mitigate risks.
- Accelerating Innovation: Data-driven insights fuel strategic decision-making, enabling rapid market adaptation.
- Enhancing Customer Experience: Intelligent agents anticipate customer needs, delivering personalized solutions.

Although the potential benefits are immense, successful implementation demands a strategic approach that encompasses ethical considerations, security, and complexity management.

Join Utkarsh Contractor, VP of AI/ML at Aisera, as we explore the opportunities and challenges of this paradigm shift and how organizations can harness agentic AI to achieve sustained competitive advantage.

James Evans, Amplitude, Director of Product Management

The past two years have been spent exploring AI’s capabilities. Now, businesses are looking to make AI investments that actually drive value by giving customers what they want, how and when they want it. Tying AI to business outcomes will be critical in 2025. At the end of the day, AI is just software. It can be incredibly powerful, but it is only valuable if it solves a customer problem. In this session, Francois James will discuss how innovators can use the right data, the right tools, and the right strategies to know exactly the AI tools to build or invest in to drive bigger business KPIs.

Anton McGonnell, SambaNova Systems, VP of Product

This session, led by Anton McGonnell, Head of Software Products at SambaNova Systems, presents an in-depth exploration of the Composition of Experts approach in AI - a model made up of many models that work in conjunction with each other - and in particular the benefits to individual developers. Instead of building one monolithic model, this strategy composes workflows using many specialised, independent "expert" models. This proves to be more efficient, modular, and extensible.

The talk will provide a deep dive into the advantages of this technique, which promotes reusability, saves on alignment tax, and enables granular access control. By detailing real-world implementations, attendees will learn best practices for breaking down problems, assigning appropriate experts, and orchestrating data flow between them. McGonnell will discuss how an architecture with a large memory capacity, enables this approach by hosting and serving thousands of models in parallel on a single node without performance degradation. This makes the Composition of Experts approach highly scalable and reduces the total cost of ownership for inference.

Additionally, the session will offer crucial hardware considerations for executing Composition of Experts at scale. It will examine how specialised architectures can allow hosting thousands of models in parallel on a single node. With companies increasingly relying on AI, reliability and cost-efficiency are paramount concerns. Composition of Experts tackles these challenges through modularity and inference optimisation. By attending, developers and engineers can expect to gain highly applicable skills for improving the scalability, performance, and governance of AI in production environments.

Dev Roundtables / Product Roundtables / AI Roundtables give attendees the opportunity to meet in small groups with DeveloperWeek + ProductWorld + AI DevWorld speakers and industry leaders to discuss real-world topics relevant to you. These open roundtables are working sessions for in-depth industry conversations.

Jocelyn Matthews, Pinecone, DevRel, Head of Community

If you had a collection of every kind of animal on earth, from mules to narwhals to goldfish, how would we pick out just the animals that are housepets? And how does AI distinguish between a pack mule and a Moscow Mule?

Embeddings are numerical representations that capture the essential features and relationships of discrete objects, like words or images, in a continuous vector space, enabling tasks such as semantic search, clustering, and recommendations.

This is a dense but entertaining talk about the core concepts of embeddings, ranging from multidimensional vector space to the ever-familiar box of crayons we used as kids. The presentation will cover fundamental concepts, including how embeddings represent various data types, the importance of context sensitivity, and between dimensionality.

The talk introduces specialized terms and breaks them down into simple, everyday examples to help learners relate difficult concepts to familiar things. Then, we build back up to the specialized terms to deepen understanding.

Attendees will gain a clear understanding of how embeddings work, their benefits, and why they are leveraged in machine learning tasks