An airhacks.fm conversation with Alvaro Hernandez ( @ahachete ) about: Framework laptop experience and build process with DIY edition, modular connectors and upgradability, running Ubuntu 25.10 beta with nix package manager, automating installation with YAML and Ansible , comparison with IBM AS/400 feature activation model, docker adoption history for server maintenance and documentation, PostgreSQL extensions, upgradability and security concerns, challenges with packing 1000+ extensions into container images, security concerns with large monolithic images containing unused extensions, dynamic extension injection using sidecar pod local controller in kubernetes , problems with mutating running containers and security tool compliance, traditional Docker build approach requiring users to become image maintainers, challenging assumptions about container image immutability and Merkle tree , container images as JSON manifests pointing to tar file layers, Dynamic OCI Registry concept for composing images on-the-fly, generating manifests dynamically in milliseconds without Docker build, interface-based approach for mapping user preferences to layer digests, PostgreSQL-specific implementation with extension URL patterns, metadata storage in PostgreSQL database for layer digest resolution, potential applications for quarkus and Java microservices, serverless deployment possibilities with AWS Lambda , comparison with Cloudflare's serverless OCI registry, enterprise use cases for automated patching and security updates, integration possibilities with AWS EventBridge for CI/CD pipelines, transparency to Docker clients with only registry change required, stackgres platform using 4 million lines of Java code, ongres company services including PostgreSQL training and Oracle migrations, Alvaro's website: aht.es Alvaro Hernandez on twitter: @ahachete…

An airhacks.fm conversation with Ronald Dehuysser ( @rdehuyss ) about: JobRunner evolution from open source to processing 1 billion jobs daily, carbon-aware job processing using European energy grid data ( ENTSO-E ) for scheduling jobs during renewable energy peaks, correlation between CO2 emissions and energy prices for cost optimization, JobRunner Pro vs Open Source features including workflows and multi-tenancy support, bytecode analysis using ASM for lambda serialization, JSON serialization for job state persistence, support for relational databases and MongoDB with potential S3 and DynamoDB integration, distributed processing with master node coordination using heartbeat mechanism, scale-to-zero architecture possibilities using AWS EventBridge Scheduler, Java performance advantages showing 35x faster than python in benchmarks, cloud migration patterns from on-premise to serverless architectures, criticism of kubernetes complexity and lift-and-shift cloud migrations, cost-driven architecture approach using AWS Lambda and S3, quarkus as fastest Java runtime for cloud deployments, infrastructure as code using AWS CDK with Java, potential WebAssembly compilation for Edge Computing , automatic retry mechanisms with exponential backoff, dashboard and monitoring capabilities, medical industry use case with critical cancer result processing, professional liability insurance for software errors, comparison with executor service for non-critical tasks, scheduled and recurring job support, carbon footprint reduction through intelligent scheduling, spot instance integration for cost optimization, simplified developer experience with single JAR deployment, automatic table creation and data source detection in Quarkus, backwards compatibility requirements for distributed nodes, future serverless edition possibilities Ronald Dehuysser on twitter: @rdehuyss…

An airhacks.fm conversation with Maurice Naftalin ( @mauricenaftalin ) about: retirement philosophy and work-life balance for developers, transitioning from paid work to passion projects, the challenge of relaxation and meditation versus constant activity, the importance of experiencing boredom in a fast-paced world, Java collections framework design and evolution over 30 years, the Collection interface as base for sets lists and queues but not Maps, mathematical foundations of sets using Zermelo-Fraenkel axioms, differences between mathematical sets and Java set implementations, NavigableSet and SortedSet using comparators versus hash-based equality, non-commutative equality between HashSet and NavigableSet implementations, CopyOnWriteArraySet for concurrent read-heavy operations with snapshot isolation, ConcurrentSkipListSet as thread-safe tree structure using skip lists algorithm, skip lists simulating tree behavior through parallel linked lists with sparse copies, Queue interface uncomfortable fit with Collection interface focusing on head/tail operations, BlockingQueue implementations for producer-consumer workflow scenarios, Deque (double-ended queue) enabling work-stealing patterns in Fork-Join framework, Map interface separate from Collection hierarchy representing key-value pairs, Map.of() factory methods using array-based optimization limited to 10 elements, WeakHashMap using weak references on keys for memory-sensitive caching, IdentityHashMap using reference equality (==) useful for serialization graphs, EnumMap and EnumSet using bitmap optimization for performance, String.intern() optimization hack for fast string comparison using reference equality, enum design limitations with final name() method preventing override customization, Epsilon garbage collector for short-lived servers avoiding GC overhead, the remarkable durability of Josh Bloch's original Collections Framework design under time pressure, balancing API simplicity with supporting unusual use cases, converting between different data structure representations and naming conventions, the Java Generics and Collections book Maurice Naftalin on twitter: @mauricenaftalin…

An airhacks.fm conversation with Ingo Kegel ( @IngoKegel ) about: jclasslib bytecode viewer development history starting in 2001, transition from CVS to Subversion to Git, SourceForge to GitHub migration, Swing UI development with FlatLaf look and feel, comparison between Swing and SWT APIs, Eclipse plugin development experiences, Visual Studio Code integration with jprofiler , Homebrew package management for Mac applications, Java desktop module and modularization, jlink for creating trimmed JDK distributions, security benefits of shipping only required modules, Java compatibility improvements since Java 17, Base64 encoder becoming public API, internal API access restrictions with module system, comparison of Java installation simplicity versus Node.js and python , potential JSON support in future JDK versions, NetBeans integration attempt and recognition issues, bytecode instrumentation for profiling, asm and ByteBuddy as standard bytecode manipulation libraries, class file format evolution and complexity, module system introducing new structures, stack map tables and verification challenges, using JClassLib for method signature extraction, dokka documentation system for Kotlin , package.md and package-info documentation patterns, potential revival of Swing for modern desktop applications, simplified application architectures compared to enterprise apps with 30-40 tabs, LLM and AI making applications simpler with chat interfaces, JClassLib use cases including learning JVM internals and editing class files, approximately 3000 GitHub stars indicating 30000+ users, IntelliJ IDEA plugin availability, physicist background influencing interest in Java internals, Java Language Specification and Class File Format books, experimental physics approach to understanding JVM Ingo Kegel on twitter: @IngoKegel…

An airhacks.fm conversation with Ingo Kegel ( @IngoKegel ) about: jprofiler Visual Studio Code integration using Kotlin Multiplatform , migrating Java code to Kotlin common code for cross-platform compatibility, transpiling to JavaScript for Node.js runtime, JClassLib bytecode viewer and manipulation library, Visual Studio Code's Language Server Protocol (LSP), profiling unit tests and performance regression testing, Java Flight Recorder (JFR) for production monitoring with custom business events, cost-driven development in cloud environments, serverless architecture with AWS Lambda and S3 , performance optimization with parallelism in single-CPU environments, integrating profiling data with LLMs for automated optimization, MCP servers for AI agent integration, Gradle and Maven build system integration, cooperative window switching between JProfiler and VS Code, memory profiling and thread analysis, comparing streams vs for-loops performance, brokk AI's Swing-based LLM development tool, context-aware performance analysis, automated code optimization with AI agents, business event correlation with low-level JVM metrics, cost estimation based on cloud API calls, quarkus for fast startup times in serverless, performance assertions in System Tests , multi-monitor development workflow support Ingo Kegel on twitter: @IngoKegel…

An airhacks.fm conversation with Stephan Janssen ( @Stephan007 ) about: Devoxx Belgium conference selling out in seconds with 7000+ page views, ticket coordination strategies by companies, VAT validation issues with European services, conference featuring largest Java AI content globally, AI and LLM focus with dedicated GenAI track, MCP and AI agents presentations, Java on GPU with tornadovm and Babylon project, quarkus AI agents for langchain4j , vibes coding experiences with React and Angular , importance of specifications for AI code generation, using multiple LLMs for review (Claude, GPT, DeepSeek), file system RAG replacing vector databases, MCP as evolution of RAG, Java as optimal language for LLM code generation due to type safety and standards, bce pattern and MicroProfile for consistent code structure, package-info files improving LLM context understanding, junior developers benefiting from AI as universal tutor, CSS generation without manual coding, dark/light theme support using CSS variables, enterprise challenges with dependency management, DevoxxGenie plugin popularity in enterprises with restricted AI access, open source contribution challenges, local LLMs becoming powerful for enterprise inference, TornadoVM Metal support enabling local inference, comparison with GraalVM inference performance, AI disrupting video production and creative industries, importance of prompt engineering skills, renaissance of development through AI empowerment Stephan Janssen on twitter: @Stephan007…

An airhacks.fm conversation with Antonio Goncalves ( @agoncal ) about: journey from Java Champion to Principal Software Engineer at Microsoft focusing on AI, the evolution from Java EE standards to modern AI development, writing technical books with LLM assistance, langchain4j as a Java SDK for LLMs providing abstraction over different AI providers, the importance of Java standards and patterns for LLM code generation, Boundary Control Entity (BCE / ECB) pattern recognition by LLMs, quarkus integration with LangChain4J enabling dependency injection and multi-tenancy, MCP (Model Context Protocol) as a new standard potentially replacing some RAG use cases, enterprise AI adoption using Azure AI Foundry and AWS Bedrock, model routers for optimal LLM selection based on prompt complexity, the future of small specialized models versus large general models, tornadovm enabling Java execution on GPUs with 6x performance improvements, GraalVM native compilation for LLM applications, the resurgence of Java EE patterns in the age of AI, using prompts as documentation in READMEs and JavaDocs, the advantage of type-safe languages like Java for LLM understanding, Microsoft's contribution to open source AI projects including LangChain4J, teaching new developers with AI assistance and the importance of curiosity, CERN's particle accelerator and its use of Java, the comparison between old "hallucinating architects" and modern LLM hallucinations, writing books about AI using AI tools for assistance, the structure of the Understanding LangChain4j book covering models RAG tools and MCP, enterprise requirements for data privacy and model training restrictions Antonio Goncalves on twitter: @agoncal…

An airhacks.fm conversation with Michalis Papadimitriou ( @mikepapadim ) about: GPU acceleration for LLMs in Java using tornadovm , evolution from CPU-bound SIMD optimizations to GPU memory management, Alfonso's original Java port of llama.cpp using SIMD and Panama Vector API achieving 10 tokens per second, TornadoVM's initial hybrid approach combining CPU vector operations with GPU matrix multiplications, memory-bound nature of LLM inference versus compute-bound traditional workloads, introduction of persist and consume API to keep data on GPU between operations, reduction of host-GPU data transfers for improved performance, comparison with native CUDA implementations and optimization strategies, JIT compilation of kernels versus static optimization in frameworks like tensorrt , using LLMs like Claude to optimize GPU kernels, building MCP servers for automated kernel optimization, European Space Agency using TornadoVM in production for simulations, upcoming Metal backend support for Apple Silicon within 6-7 months, planned support for additional models including Mistral and gemma , potential for distributed inference across multiple GPUs, comparison with python and C++ implementations achieving near-native performance, modular architecture supporting OpenCL PTX and future hardware accelerators, challenges of new GPU hardware vendors like tenstorrent focusing on software ecosystem, planned quarkus and langchain4j integration demonstrations Michalis Papadimitriou on twitter: @mikepapadim…

An airhacks.fm conversation with Adam Dudczak ( @maneo ) about: early programming experiences with Commodore 64 and Pascal , demo scene participation through postal mail swapping of floppy disks, writing assembly code for 64K intros with music and graphics, developing digital library systems using Java Servlets and Hibernate , involvement in reactivating Poznan Java User Group in 2007, NetBeans Dream Team and NetBeans World Tour, appearing on Polish breakfast TV to discuss Java programming, working at Supercomputing Center on cultural heritage digitization projects, transitioning to EJB 3.0 and Glassfish based on conference inspirations, joining allegro in 2014 to rewrite search functionality from PHP to Java microservices, handling 14K requests per second with Solr-based search infrastructure, migrating big data stack from on-premise Hadoop to Google Cloud Platform, developing private banking application for children using Spring and Hibernate then migrating to Google Sheets with 70 lines of JavaScript, discussing public cloud cost optimization strategies, comparing AWS Lambda versus EC2 versus container services based on traffic patterns, emphasizing removal of code when moving to public cloud to leverage managed services, standardization benefits of Java EE for long-term maintenance and migration, quarkus as modern framework supporting old Jakarta EE code with fast startup times, importance of choosing appropriate persistence layer (S3 vs relational databases) based on cloud costs, serverless architectures for enterprise applications with predictable low traffic, differences between AWS Azure and GCP service offerings and pricing models, Turbo assembler project klatwa Adam Dudczak on twitter: @maneo…

An airhacks.fm conversation with Maurice Naftalin ( @mauricenaftalin ) about: Java generics history and introduction timeline, the Pizza paper proposal by Phil Wadler and others, Type erasure vs Reification debate, binary compatibility considerations, Java Community Process politics and Apache Harmony licensing issues, evolution from JCP to JEPs process, legacy collection types like Vector and Hashtable with small 't' naming inconsistency, thread safety removal in Java 2 collections, generics applied to classes methods and constructors, unusual generic constructor example, Covariance and contravariance , invariant covariant and contravariant types, array covariance and ArrayStoreException, wildcards with extends and super bounds, PECS (Producer Extends Consumer Super) principle by Josh Bloch, Get and Put principle alternative, sealed interfaces potential impact on generics, reflection access to generic type information despite erasure, Class generification and type literals, raw types vs unbounded wildcards distinction, reifiable types definition, unchecked warnings importance, Service Loader usage with generic type checking for plugin systems, minimalist Java SE approach for long-lasting enterprise projects, syntactic sugar as code generation, records implementation as constrained classes, comparison of C++ templates criticism to Java generics adoption, the Java Generics and Collections book Maurice Naftalin on twitter: @mauricenaftalin…

An airhacks.fm conversation with Jonathan Ellis ( @spyced ) about: brokk as a Norse dwarf who forged Thor's hammer, Java Swing UI performance advantages over Electron apps, zb build tool integration, onboarding experience comparison with Cursor, architect vs code buttons functionality, session management in brokk, build and test tool configuration, in-memory Java parser development, JVector and embedding models limitations, agentic search approach using find symbol by wildcard and fetch method tools, hierarchical embeddings concept, package-info for AI context, LLMs as artists needing constraints, Java's typing system advantages for AI feedback, architect mode with multiple tool access, code agent feedback loops, joern code graph indexing, Git integration with jgit , custom diff format avoiding JSON escaping issues, tool calling in architect mode, MCP server development in pure Java - zmcp , prompt templates for team collaboration, JBang installation experience, subscription pricing discussion, organizational subscriptions for corporate teams, avoiding context explosion in architect mode, Gemini Flash for summarization, workspace tools and summaries, build status feedback to architect, enterprise-friendly features development Jonathan Ellis on twitter: @spyced…

An airhacks.fm conversation with Fabio Niephaus ( @fniephaus ) about: GraalVM polyglot capabilities now available as Maven dependencies without requiring GraalVM JDK, running WebAssembly modules in Java applications using GraalWasm , separation of polyglot runtime from GraalVM distribution, embedding use cases for extending Java applications with python JavaScript and WebAssembly, performance benefits when running on GraalVM vs openJDK through automatic JIT optimization, WebAssembly as portable compilation target for multiple languages including rust C++ Go, WASI (WebAssembly System Interface) enabling file and network operations, advantages over JNI/Panama FFI for native extensions due to portability and sandboxing, multi-threading support with context pools for high throughput, using JavaScript bindings as intermediary for high-level Java-WASM interactions, future component model with WIT (WebAssembly Interface Types) for language-agnostic interfaces, security benefits of sandboxed execution for untrusted code, WebImage preview feature compiling Java bytecode to WebAssembly modules, javac demo running Java compiler in browser, command-line tools converted to web applications using WebImage, Edge Computing use cases for user-defined functions, native image compatibility with GraalWasm, Pyodide integration possibilities for secure Python native extensions, Spring Shell successfully compiled to WASM demonstrating framework compatibility, ongoing work on threading networking and WASI support for full server-side capabilities, collaboration with WebAssembly community and Bytecode Alliance, WASM GC proposal for efficient garbage collection, bringing dynamic class loading to native image, GraalWasm demos and guides , javac on Wasm live demo , javac on Wasm demo code , Web Image talk at Wasm.io 2025 , GraalVM Web Image sources , GDK Launcher , GraalPy , GraalPy demos and guides Fabio Niephaus on twitter: @fniephaus…

An airhacks.fm conversation with Maurice Naftalin ( @mauricenaftalin ) about: experiences with Visual Age for Java and its visual programming approach with arrows connecting components, working on British Department of Health and Social Security project using Visual Age for Java for benefits system navigation, comparison of various Java IDEs including Visual J++, Sun Java Workshop , JBuilder , Eclipse , NetBeans , IntelliJ IDEA, and Visual Studio Code , advantages of VS Code for polyglot programming and its growing ecosystem, visual programming experiences with state charts for reactive systems, IBM Rational tools and UML integration, successful visual programming with NetBeans Matisse GUI builder and AWS Step Functions , Model Driven Architecture and code generation from UML diagrams, writing Java Generics and Collections book with Philip Wadler for Java 5 and updating it for a second edition, changes in Java idioms over 15 years including deprecation of wrapper class constructors, sequence collections as major addition to Java collections framework, PECS (Producer Extends Consumer Super) principle for generics, underappreciated Java collections like NavigableMap, preference for method references and keeping lambdas concise in streams, using Class::method notation instead of Class.method, Scottish countryside and Edinburgh living experiences, early internet challenges with 300 baud acoustic couplers influencing views on network distribution versus CD-ROMs, transition from safety-critical systems to Java training and consulting, importance of understanding bounded wildcards in generics, future impact of Project Valhalla on generics and collections Maurice Naftalin on twitter: @mauricenaftalin…

An airhacks.fm conversation with Michalis Papadimitriou ( @mikepapadim ) about: starting with Java 8, first computer experiences with Pentium 2, doom 2 and Microsoft Paint , university introduction to Object-oriented programming using Objects First and bluej IDE, Monte Carlo simulations for financial portfolio optimization in Java, porting Java applications to OpenCL for GPU acceleration achieving 20x speedup, working at Huawei on GPU hardware, writing unit tests as introduction to TornadoVM , working on FPGA integration and Graal compiler optimizations, experience at OctoAI startup doing AI compiler optimizations for TensorFlow and PyTorch models, understanding model formats evolution from ONNX to GGUF , standardization of LLM inference through Llama models, implementing GPU-accelerated Llama 3 inference in pure Java using TornadoVM, achieving 3-6x speedup over CPU implementations, supporting multiple models including Mistral and working on qwen 3 and deepseek , differences between models mainly in normalization layers, GGUF becoming quasi-standard for LLM model distribution, TornadoVM's Consume and Persist API for optimizing GPU data transfers, challenges with OpenCL deprecation on macOS and plans for Metal backend, importance of developer experience and avoiding python dependencies for Java projects, runtime and compiler optimizations for GPU inference, kernel fusion techniques, upcoming integration with langchain4j , potential of Java ecosystem with Graal VM and Project Panama FFM for high-performance inference, advantages of Java's multi-threading capabilities for inference workloads Michalis Papadimitriou on twitter: @mikepapadim…

An airhacks.fm conversation with Jonathan Ellis ( @spyced ) about: brokk AI tool for code generation named after Norse god of the forge, AI as complement to experienced programmers' skillsets, age and productivity in programming, transition from JVector to working on Cassandra codebase, challenges with AI in large codebases with extensive context, building tools for historical Java codebases, comparison of productivity between younger and older programmers, brute force coding vs experienced approach, reading code quickly as a senior skill, AI generating nested if-else statements vs better structures, context sculpting in Brokk, open source nature of Brokk, no black boxes philosophy, surfacing AI context to users, automatic context pulling with manual override options, importing dependencies and decompiling JARs for context, syntax tree based summarization, Maven and Gradle dependency handling, unique Java-specific features, multiple AI model support simultaneously, Claude vs Gemini Pro performance differences, Git history as context source, capturing commits and diffs for regression analysis, migration analysis between commits, AI code review and technical debt cleanup, style for code style guidelines, using modern Java features like var and Streams, Error Prone and NullAway integration for code quality, comparison with Cursor's primitive features, branching conversation history, 80% time in Brokk vs 20% in IntelliJ workflow, sketching package structures for AI guidance, data structures guiding algorithms, Git browser by file and commit, unified diff as context, reflection moving away from due to tooling opacity, Jackson serialization refactoring with DTOs, enterprise features like session sync and sharing, unified API key management, rate limit advantages, parallel file processing with upgrade agent, LiteLLM integration for custom models, pricing model based on credits not requests, $20/month subscription with credits, free tier models like Grok 3 Mini and DeepSeek V3, architect mode for autonomous code generation, code button for smaller problems with compile-test loop, ask button for planning complex implementations, senior vs junior programmer AI effectiveness, self-editing capability achieved early in development, no vector search usage despite JVector background Jonathan Ellis on twitter: @spyced…