Find partners
airhacks.fm podcast with adam bien

airhacks.fm podcast with adam bien

Hosted by Adam Bien

TechnologyInterviews guestsExplicit

Episodes

421

Latest episode

Nov 2026

Language

EN-US

About the show

Java, Serverless, Clouds, Architecture and Web conversations with Adam Bien

Listen to episodes

60 recent
November 19, 2026Episode 4221 hr 4 min

About Event Sourcing

An airhacks.fm conversation with Johan Haleby about: returning guest Johan Haleby, Midsummer pole dancing in Sweden, domain events as facts and decisions that cannot be undone, domain events understandable by business people, event sourcing as storing a list of facts instead of the current state, deriving state from events to make new decisions, storing events vs event sourcing, snapshotting as a technical optimization applied only when needed, closing the books as a domain oriented alternative to snapshotting, tractors returning to the garage as a natural end of a business session, streams per aggregate and reading only the events needed for a decision, commands, deciders and aggregates enforcing business invariants, consistency per farm instead of global consistency, Kafka Streams cascading low-level GPS events into speed, direction and fuel consumption events, translating technical events into business events, Kafka Streams joins and tables on demand as a reversed database, AWS Lambda durable functions as primitive event sourcing per stream, low-level UI events like mouse moved vs business events like player joined game, event sourcing per bounded context, reactive programming vs event sourcing streams, CQRS as an optional addition to event sourcing, read your own writes and Eventual Consistency of read models, BASE vs ACID, a last updated timestamp as a business solution for stale views, writing to PostgreSQL and reading from Elasticsearch , optimistic locking for concurrent decisions, write side vs read side and subscriptions to committed events, Kafka as event streaming platform, not an event store, in-memory event sourcing without distribution, IoT battery state of charge events and notification deciders, Java Flight Recorder events as an observability use case, bank accounts and state transitions as the classic event sourcing use case, events as stories easier to reason about than table relationships, decoupling and extensibility with events, replaying Kafka topics for new microservice versions, storing events in Amazon DynamoDB or with Amazon S3 conditional writes and consistency guarantees per aggregate, events generated by deciders rather than by commands, a future episode about the Occurrent library Johan Haleby on twitter: @johanhaleby

November 12, 2026Episode 4211 hr 14 min

How Behaviour-Driven Development (BDD) Happened

An airhacks.fm conversation with Daniel Terhorst-North about: ThoughtWorks in 2003 and XP technical practices, test-driven development, continuous integration and pair programming, learning an instrument and team sports as metaphors for practice, learning, practice and training as first class Kaizen work, critics of a practice becoming its advocates, Virginia Satir and the principle of positive intent, sabotaged projects and "if you reward firefighters, you breed arsonists", the curved tables pair programming story, agile methods at snowbird started as small teams in big companies, pair programming as two people solving a problem rather than driver and navigator, the 2003 Java Developers Journal article "test driven development is not about testing", introducing TDD to a team without using the word test, tests as executable requirements written in the implementation language, discovering domain terms and Domain-Driven Design, example guided design, Kent Beck: "as few tests as I can get away with", single examples as seeds for property-based testing, main methods as test drivers before JUnit , zb zero dependency builder for Java 25 , zunit test runner with Java assert and process isolation, Daniel Worthington Bodart's 10 second build and shavenmaven , Apache Ivy and dependency lock files, tinytest single header C test framework, zsmith agent harness with Java tools and lightmetal inference, migrating GitHub Spec Kit from python , jq and shell scripts to Java 25, Chris Matts and analysis as specification, Ivan Moore's index cards with "I do this, this happens", given when then, Hoare logic and arrange act assert, scenarios named like Friends episodes, rbehave and RSpec describe and it inspired by HyperCard, self-documenting colored scenario output, the plain text story runner, Aslak Hellesøy's Cucumber with a Treetop grammar, Gherkin from pickling a cucumber, criticism of BDD for getters and setters, interaction testing with Mockito as double entry bookkeeping, FitNesse and Excel test data with Apache POI HSSF, EARS requirements syntax from Rolls-Royce for spec-driven development, BDD adoption over 20 years, LLM guidance as heuristics vs rules enforced outside the LLM, Gojko Adzic's observations to heuristics to rules to lint rules, Claude Code hooks, rules and CLAUDE.md, switching between Claude Code, Codex and Mistral Vibe CLI , skills documenting decisions such as preferring records, Jakarta EE and MicroProfile as a constrained API set, Boundary Control Entity and ArchUnit , probing LLM knowledge in a toolless playground, context drift and short rule files, a Marp based class builder and teacher heads up display built with Claude, in-person training and consulting via Goalwards Daniel Terhorst-North on bluesky: @tastapod.com

November 6, 2026Episode 4201 hr 5 min

CDI Boot Before Quarkus

An airhacks.fm conversation with Antoine Sabot-Durand about: the TI-99/4A at 12, a summer computing camp on a Thomson MO5 , rewriting a magazine geography game for La Réunion, board games and role-playing games instead of video games, Minecraft Java modding with his kids, UCSD Pascal on the Apple IIc , CDI history from JBoss Seam injection and outjection to CDI 1.2 and CDI 2.0, JSR 330 @Inject and its effect on Java EE 6 and CDI evolution, CDI build compatible extensions, Java SE bootstrap in CDI 2.0, the CDI boot initiative vs WildFly Swarm and Thorntail, class loading issues in shrunk application servers, MicroProfile Health and MicroProfile Fault Tolerance , Quarkus adoption in a large European bank with Open Liberty as alternative, standards for upgradability, Quarkus on AWS Lambda and Azure Functions , MicroProfile API surface instead of Quarkus extensions, javax to jakarta namespace migration, Jakarta EE and MicroProfile specifications and TCKs for the LLM era, saving inference costs, one-shot accuracy without hallucinations, Vidocq as a cliffhanger Antoine Sabot-Durand on twitter: @antoine_sd

October 27, 2026Episode 4191 hr 1 min

A Billion Data Points Per Second

An airhacks.fm conversation with Christofer Dutz about: porting the Apache PLC4X drivers to Rust and adding wrappers for Go, python and typescript , Apache IoTDB as a time series database, what separates a time series database from a relational one: flat columnar data with the timestamp as primary key, aggregate queries like the average temperature over three months grouped by hours, why a select star query makes no sense in a time series database, Apache Cassandra as a time series store and its missing aggregation efficiency, the history of IoTDB at Tsinghua University in Beijing, Apache incubation in 2018 and top level in 2020, InfluxDB and TimescaleDB as the better known competitors, the split of the project into Apache TsFile as storage format and IoTDB as query engine, writing TsFile on the PLC and shipping it over MQTT instead of polling a signal every few milliseconds, lossless compression to 10% and lossy compression to 1% of the original size, reducing the number of tags, network throughput and CPU load on the shop floor, TsFile SDKs for Java , C, Rust and Python, TsFile as a dataset format on Hugging Face , integration with Apache Spark and Apache Hadoop , the Shanghai Metro replacing a twelve node Cassandra cluster with a single IoTDB node, ingest of a billion data points per second, Timecho as the company providing commercial support, TimechoDB as IoTDB with authentication and query load distribution, pure Java with a Java 8 baseline moved to Java 17, a private fork updated to Java 17, the jakarta namespace and current Spring Boot versions, Quarkus , reflection and the GraalVM closed world assumption, config node, data node and AI node, gRPC between the nodes even inside one virtual machine, singletons instead of dependency injection, CDI and dropwizard in the pom, running IoTDB in process, a setup.exe install compared to a Cassandra cluster with docker and kubernetes , starting an inference process in process instead of over HTTP, lightmetal and zsmith as an example, Vidocq as a CDI implementation without external dependencies, the JDBC driver , the REST API and an Apache Drill plug-in, push queries that call back whenever the query produces a new result, hierarchies of aggregation levels with per level retention from milliseconds on the shop floor to years for business analysis, the tree identifier of company, site, line, machine and machine part, the table model added for users who think in table algebra, data types, missing updates, delete and automatic purging by retention, replication so queries run on a second instance, the built-in MCP server giving Claude access to an IoTDB instance, the AI node running models inside the database, data gravity, a KNX house logging into IoTDB on a Raspberry Pi , an IoTDB app running in process on a Bosch Rexroth ctrlX PLC Christofer Dutz on linkedin: christofer-dutz

October 13, 2026Episode 418

Agents Have No Runtime Insight

An airhacks.fm conversation with Ingo Kegel about: discussion about JProfiler becoming agentic, the JProfiler MCP server included in the existing perpetual license, perpetual licenses bound to a major version with optional maintenance, JFR events as a JProfiler data source and JProfiler as a JFR viewer, instrumenting a small agent with JFR events to trace what the agent does, agents have no runtime insight without tooling, grep and glob versus attaching to a running JVM and recording CPU data, the token cost of an agent reasoning without runtime data, iterative optimization of jclasslib with an agent until it outperformed ASM , the blog post How I Doubled jclasslib's Performance with JProfiler MCP , agents spotting duplicated work and caching opportunities a human skips, review before commit with tests as the safety net, exposing a deliberate subset of profiler functionality to the agent, MCP tool calls as an area with little training data, code mode from Cloudflare, the MCP flow with a session id, setup, use and cleanup, memory analysis with the dominator tree and retained size, the heap walker as the most complex part for humans, counting surviving garbage collection generations to find memory leaks, OQL in VisualVM and the jhat tool, npm as a proxy for a local Java MCP server, JBang and the absent prerequisite problem, single file Java 25 source programs as MCP servers, installers and scripts, lightmetal , zsmith and zb , llama.cpp behind the FFM API and TornadoVM for pure Java inference, Gemma models for local one shot automation, ServiceLoader with BinaryOperator to couple two projects without shared code, hard wiring tools and skills into task specific agents instead of discovery, MCP as a facade over tools that a custom agent does not need, conflicting skills pulling an agent in different directions, the install4j skill and its evaluation harness with external judges, the AI probe in JProfiler 16 measuring input and output tokens per call tree, prompt to token ratio as a metric, normative APIs and Java being well represented in LLM training data, Boundary Control Entity packages named after the domain, Ivar Jacobson and the age of the pattern, AI Rails skills, Claude Code and Codex , continuous testing and continuous performance testing with an exit condition, zb as a zero dependency builder with single file tests using assert , the missing JSON API in the JDK , where to find JProfiler and ej-technologies Ingo Kegel on twitter: @IngoKegel

October 13, 2026Episode 4171 hr 13 min

Incidental Architecture

An airhacks.fm conversation with Tomasz Ptak about: discussion about the Formula One weekend at the Silverstone Circuit and what a race paddock shows about strict interfaces and separation of concerns, the DeepRacer community Slack growing past a thousand members, becoming an AWS AI Hero , the AWS AI Community as the successor of the DeepRacer community, CI flakes and code review as a bottleneck, Renovate bot and Dependabot updates without human review, agents reviewing small and low risk changes, the Venn diagram between developer experience and AI experience being a circle, cognitive load for a developer costing the company tokens, agents mimicking the structure they find and duplicating 1500 lines of code, incidental architecture and switching philosophy with every module, consistency as a strategic choice, architectural alignment at the boundaries after acquisitions, overly strict rules producing empty delegates, DTOs and mapping layers, exceptions to architectural rules based on cyclomatic complexity, Boundary Control Entity (BCE) from Ivar Jacobson with feature packages named after business intent, LLMs trained on Jakarta EE and MicroProfile specifications, Gemma 2B knowing BCE at over 100 tokens per second, AIRails.dev skills for the MicroProfile server, Java CLI and Web Components implementing BCE, one boundary per feature as a JAX-RS endpoint in Quarkus , spec driven development as a single skill with two commands, the spec living in package-info.java as Markdown Javadoc , jmarkdoc exporting JavaDoc as Markdown documents, EARS requirements syntax and one test per requirement, unit, integration and System Tests as the verification loop, code as a cache of the understanding of the requirements, FinOps and token cost as the reason to use what the model already knows, grep and find versus the Language Server Protocol (LSP) in Claude Code , javap decompiling jar dependencies and losing context, keeping module sources visible to the agent instead of hiding them in jars, monorepo and monobuild trade-offs, coarse grained modules instead of hundreds of microservices, microservice architecture for agents, zb as a zero dependency Java builder, quality as an amplifier in both directions, a library carrying an instruction that tells agents to delete the tests, trust and verification instead of reading every line, Kiro with one spec per business component and its requirements, inference cost as the question from the GeeCON audience, the sustainability point beyond which agents amplify maintainability Tomasz Ptak on linkedin: tomasz-ptak

October 7, 2026Episode 4161 hr 18 min

"my favorite framework is void main()"

An airhacks.fm conversation with Andrzej Grzesik about: text editors as the first encounter with persistence, load and save in Basic and later Turbo Pascal , the question why only one file is startable, editing save games with a hex to decimal calculator from a games magazine, peek and poke commands and the memory map of early home computers, BAT files and prepared game installers, dBASE , Norton Commander and Midnight Commander , a Linux account in a Krakow secondary school, operator overloading and implementing RSA in C++ , a multimedia course covering JPEG compression and Runge-Kutta, Pascal and Delphi with the Borland Visual Component Library, first contact with Java on the command line, the JVM promise against hardware specific bugs, the Sun Campus Ambassador program , Sun as a high tech company with Jini , JavaSpaces , JXTA , Apache Derby and OpenSolaris , OpenSolaris boot environments and ZFS snapshots as a predecessor of containers and the NixOS parallel, the first serious Java project at Sabre: airline software on JDK 1.4 with Enterprise Beans targeting Solaris, ClearCase , setting up an office in Krakow for e-Circle, Apache Hadoop and Apache HBase around 2010 before Apache Cassandra , picking patches from Apache Subversion and hoping the result compiles, Groovy and GPars for concurrency, learning Puppet and Chef to run a Hadoop cluster in VMS , Hadoop API verbosity compared to Enterprise Beans, finding and deleting a hundred duplicated DAO classes, GlassFish as the reference implementation with commercial support and live redeploy from the command line, web framework fatigue and blog posts that do not reproduce, Apache NetBeans and EJBs for a desktop client with network communication, one stateless annotation per class as the essential EJB style, eBay, customer data analytics at Burberry, replacing overnight batches with record at a time processing, the can you be faster than a MacBook competition against a ten machine Hadoop cluster, head of backend at Revolut and PostgreSQL for transactions, LMAX, the LMAX disruptor , counting microseconds, public static void main used as the framework and constructor arguments that stay visible, founding DataLinks , teaching the computer what related means, ontologies, RDF and SPARQL , defining flexible relations between tables and walking them as a graph instead of materializing them, neurosymbolic AI , connecting sanction lists from PDF, Excel and XML, JavaOne then and now, GeeCON , JDD , infoShare and Code Europe Andrzej Grzesik on twitter: @ags313

September 29, 2026Episode 4151 hr 13 min

select, reject, detect, collect and inject

An airhacks.fm conversation with Donald Raab about: the relation between Clipper and FoxPro , Smalltalk blocks with Square brackets and a vertical pipe separating the parameters, the knights of the square bracket, the pipe replaced by the arrow in Java , block context template as the single type for every SmallTalk block, anonymous inner classes used as block substitutes before Java 8, the Smalltalk hierarchy with object as a subclass of nil, nil as an instance of the undefined object class that still responds to messages, a blog post about null as an object in Java, metaclasses and the endless class of the class chain, nominal types and functional interfaces versus one type for all blocks , single abstract method (SAM) and the disappearance of the term, the combinatorial explosion of function types over primitives and generics , naming generic type parameters T and V, single character type variables and collisions with variable names, var and local variable type inference in everyday Java, var for expected values in unit tests, postfix completion in IntelliJ IDEA with dot var and dot out, changing a running Smalltalk image and porting IntelliJ IDEA key bindings back to VisualAge for Smalltalk , control Y to delete a line in WordStar and in IntelliJ IDEA, JetBrains MPS and custom DSLs, Clipper code blocks with curly braces and double square brackets around the parameters, blocks for sorting and for T-Browse columns in Clipper, Clipper visual objects compiling to native Windows executables, the Classy library for classes without inheritance in Clipper, the Smalltalk collection protocol select, collect, reject, detect and inject into, filter as an ambiguous method name compared to select and reject, transforming behaviours map, flat map and collect, String.transform in Java, the missing reject in the Java Streams API , Alice's Restaurant by Arlo Guthrie as the claimed origin of the Smalltalk method names, Smalltalk popularity in Europe and Argentina, Newspeak , OS/2 Warp and REXX , duck typing and does not understand errors in Smalltalk, Ruby as a Smalltalk dialect with select, reject and inject, asciidoctorj running on JRuby , blocks versus closures and captured state, effectively final captured variables in Java lambda expressions , Groovy closures and Grails before Java 8, the joint Groovy and Java compiler, Apache Groovy and the Gradle DSL, blog exchanges with Groovy committer Paul King , the Scala and Kotlin adoption waves, Java 25 shipping enough out of the box, the origins of Eclipse Collections in 2004 as Caramel (collection array map enumeration library) at Goldman Sachs, open sourcing as GS Collections to share use cases with the JSR 335 expert group, the ten year quest to get lambdas into Java, the BGGA prototype, lambdas in the JSR 341 expression language before Java 8, memory consumption and code duplication rather than performance as the motivation for Eclipse Collections, procedure instead of consumer and extending consumer for JDK interoperability, select with and predicate two to keep lambdas stateless and static, method references preferred over multi line lambdas, refactoring anonymous inner classes to lambdas for 10 to 20 percent less code, the commercial JGL collections framework, early access testing in the OpenJDK quality outreach program , a new Collectors.toList instance created on every call, Brian Goetz and the constraints of the Java language architect, composing records, patterns and switch expressions into data oriented programming, the book Eclipse Collections Categorically and its method categories, custom folding regions in IntelliJ IDEA to simulate method categories Donald Raab on twitter: @TheDonRaab

September 21, 2026Episode 4141 hr 4 min

Static as a Roq

An airhacks.fm conversation with Andy Damevin about: Texas Instruments TI-BASIC calculator with GOTO and GOSUB, programming games including a trading game on the calculator, sending programs from a computer to the calculator, building viral Facebook games with the social graph API, PHP and MySQL with iframes and OAuth for Facebook authentication, earning revenue with ads and paid credits, a first serious Java project simulating cars moving across lab screens over TCP/IP sockets with threads and shortest-path routing, comparison with Java mobile agents and Aglets , JBoss Seam and Java EE , moving to Spring then quarkus , founding a company and building the Wazzam events app with Java, Spring, Angular and PhoneGap , joining Red Hat and the Quarkus team, building the code.quarkus.io launcher, creating Quarkus Quinoa to integrate NPM front-end builds into the Quarkus lifecycle with a dev proxy, the Quarkus Web Bundler for bundling web applications without NPM using WebJars and Maven NPM dependencies, minification and hashed asset versioning, Deno as a faster and lighter Node alternative delivered as architecture-dependent Maven and Gradle dependencies, Roq the Quarkus static site generator exporting a running Quarkus application as HTML snapshots, the Qute template engine with compiled type-safe templates CDI access and iteration, static site publishing via Git to GitHub Pages or Netlify, using a scheduler and CDI events or jgit to regenerate and push pages offline, the RoqGenerator application-scoped bean with a generate method, dev mode versus prod mode differences, Roq as suitable for developers and AI with MCP integration and Playwright for iterative design, AI for processing content into summaries and translations from Markdown asciidoc JSON and YAML , an integrated WordPress-style editor for writers with Git sync, and devoured.fyi a static AI and developer NeWS site built with Roq using Tailwind and plain Vanilla JS Andy Damevin on twitter: @ia3andy

September 14, 2026Episode 41358 min

From the One Billion Row Challenge to a Multi-Threaded Parquet Library

An airhacks.fm conversation with Gunnar Morling about: returning guest discussion four years after episode "#174 Kafka Connect CLI, JFR Unit, OSS Archetypes and JPMS" , the One Billion Row Challenge (1BRC) as a Java community coding challenge, aggregating a 13 gigabyte semicolon separated file of one billion temperature measurements per weather station, baseline single threaded idiomatic Java at five minutes versus a 1.5 second winning entry, two orders of magnitude speedup, parallel processing scaling near linearly across cores, memory mapped files via the Foreign Function and Memory API for arbitrarily large files, reading eight bytes at once into a long, SIMD parsing via the incubating Vector API , GraalVM native binaries , no third party libraries and single source file rule, results on par with C, Thomas Würthinger and the GraalVM team participation, HTMLDB a Java 25 executable shebang script storing records as semantic HTML pages using the built in XML parser, agents reading XML better than JSON , Hardwood a minimal dependency multi threaded Java library and CLI for reading and writing Apache Parquet files, columnar versus row based formats, efficient encodings including delta encoding and bit packing, trading CPU for disk size, range requests to read specific columns from object storage like S3 , the dependency heavy Parquet Java library pulling in the Hadoop stack, zero mandatory dependencies with optional modules, a hand written AWS Signature Version 4 signer using the JDK HTTP client instead of the AWS SDK , verifying the signer against the AWS TCK test suite, owning code with LLMs, optional dependency for IAM instance profiles , Zstandard and Snappy compression as domain specific dependencies, Java LLM code generation quality and import handling, AIRails.dev skills based on Boundary Control Entity , spec driven development storing the spec in package-info Java Doc using EARS notation , the code review pyramid as a review skill, GraalVM WebAssembly web image compiling a Java text user interface to analyze Parquet files in the browser, an agent skill for Parquet analysis in natural language, Apache Iceberg and S3 Tables , Apache Flink integration, the shift in the reuse versus own trade-off with LLMs, Agent zSmith a zero dependency Java agent harness, var over explicit types Gunnar Morling on twitter: @gunnarmorling

Is this your show?

Claim this listing to keep it up to date, reach guests who want to pitch you, and manage bookings with Guestify.

Claim this listing

More Technology podcasts