Episode 105 of The Technical Co-Founder Podcast dives into the world of conflict-free replicated data types, or CRDTs. Lucas and Luna explore how a startup called Pixelboard built a real-time collaborative whiteboard that works seamlessly offline, then syncs when connectivity returns. The episode focuses on their specific engineering choices: why they chose an RGA-based list CRDT for drawing strokes, how they handle undo across peers, and the trade-off between causal delivery and performance. Lucas explains why CRDTs beat Operational Transform for this use case, citing data transfer of just 2.4 KB per stroke even with hundreds of collaborators. The hosts also touch on the operational complexity of garbage-collecting tombstones in an append-only log. This is a concrete case study for any engineering leader evaluating offline-first architecture.

#CRDT #OfflineFirst #CollaborativeEditing #RealTimeSync #Pixelboard #RGA #DistributedSystems #ConflictResolution #LocalFirst #WebDevelopment #StartupEngineering #ClientSideLogic #UndoRedo #BusinessAndTechnology #EngineeringPodcast #FexingoBusiness #BusinessPodcast #TechPodcast

Keep every episode free: buymeacoffee.com/fexingo

Episode 104 dives into the engineering trade-off between gRPC and Cap'n Proto, a zero-copy serialization format that one startup used to cut inter-service latency by 60 percent. Lucas walks through how the company swapped protobuf-generated gRPC stubs for raw Cap'n Proto messages over TCP, eliminating serialization overhead in their data pipeline. Luna presses on the costs: loss of HTTP/2 streaming, no built-in health checking, and a steeper learning curve for the team. The episode includes concrete benchmarks—300 microseconds per call down from 1.2 milliseconds—and a candid discussion of when the trade-off makes sense (high-throughput, latency-sensitive paths) versus when it doesn't (public-facing APIs or polyglot environments). The hosts also touch on the broader trend of zero-copy serialization in systems programming and why startups with tight latency budgets are revisiting old assumptions about gRPC as a default.

#CapnProto #gRPC #Serialization #ZeroCopy #Latency #InterServiceCommunication #StartupEngineering #SystemsProgramming #ProtocolBuffers #TCP #Performance #Microservices #Backend #DataPipeline #BusinessAndTechnology #FexingoBusiness #BusinessPodcast #TheTechnicalCoFounder

Keep every episode free: buymeacoffee.com/fexingo

In this episode of The Technical Co-Founder Podcast, Lucas and Luna dive into a single, sharp engineering practice: how one startup, a fast-growing fintech called Veridian, sets tail latency targets at the 99.9th percentile — aiming for under 10 milliseconds — on every critical microservice. They explain why traditional average latency metrics hide customer-impacting slowdowns, how Veridian uses a p99.9 SLO to force architectural decisions like request hedging and off-peak batch processing, and what happened when they found a service that spiked to 200ms during a flash sale. Lucas walks through the math: a single 200ms outlier can block a user from checking out, and with 50 upstream calls per request, the probability of at least one slow call skyrockets. Luna challenges whether this level of precision is worth the engineering overhead for smaller teams. They also discuss monitoring tools like histograms in Prometheus and the cultural shift from 'the system is up' to 'the system is fast enough for every user.' A concrete look at how one team uses tail latency as a product-level lever, not just a benchmark.

#TailLatency #p99 #Microservices #SLO #Veridian #Fintech #LatencyTargets #Prometheus #RequestHedging #SystemDesign #EngineeringCulture #Observability #Business #Technology #FexingoBusiness #BusinessPodcast #TechnicalCoFounder #StartupEngineering

Keep every episode free: buymeacoffee.com/fexingo

Lucas and Luna explore how a DeFi startup called VeriFi uses formal verification—mathematical proof—to guarantee smart contract correctness, avoiding exploits that cost billions in crypto. They discuss the trade-offs of proof-based development vs testing, how VeriFi caught a critical bug before launch, and why formal methods are trickling into traditional finance. With VeriFi's CTO as a case study, they unpack the practical challenges of building formally verified systems: the steep learning curve, tooling maturity, and the cost of proving simple contracts 10x more expensive than writing them. The hosts also touch on how this approach compares to fuzzing and property-based testing, and why some VCs are now mandating formal verification for core DeFi protocols.

#FormalVerification #SmartContracts #DeFi #VeriFi #Blockchain #CryptoSecurity #ProofBasedDevelopment #Solidity #BusinessAndTechnology #FexingoBusiness #BusinessPodcast #TechnicalCoFounder #EngineeringLeadership #CTO #SecureDevelopment #MathProofs #BugPrevention #ZeroExploits

Keep every episode free: buymeacoffee.com/fexingo

Most teams run unit tests and think they're covered. But unit tests only check examples you think of. Property-based testing flips the model: you define invariants and let the computer generate thousands of random inputs to try to break them. Lucas and Luna examine how one startup — a fintech payments platform called PaySignals — adopted property-based testing after a production bug caused silent rounding errors on 0.3% of transactions over three months. The bug was invisible to every hand-written test. They walk through how PaySignals used a Haskell-style QuickCheck library ported to Go, how they defined properties like 'every valid payment request maps to a non-negative fee', and what happened when they ran 50,000 random test cases and found 11 hidden edge cases in a codebase that already had 87% line coverage. They also discuss the trade-offs: slow test suites, the learning curve for developers who think in examples, and when property-based testing is worth the complexity vs. when it's overkill. This episode is for engineers who want to test smarter, not harder.

#PropertyBasedTesting #QuickCheck #GoLang #Fintech #PaySignals #EdgeCases #SoftwareTesting #RandomTesting #Invariants #QualityAssurance #DevTools #StartupEngineering #BusinessAndTechnology #FexingoBusiness #BusinessPodcast #TechnicalCoFounder #CodeQuality #TestingStrategy

Keep every episode free: buymeacoffee.com/fexingo

In this milestone 100th episode, Lucas and Luna explore how a startup called Prisma-like-but-not-Prisma is using WebAssembly to run database migrations directly in the browser. They break down why traditional migration tooling is painful, how WASM eliminates the dependency on local runtime environments, and what this means for developer onboarding and CI/CD pipelines. Lucas walks through the specific trick of compiling the migration engine to WASM and hosting it on a CDN, while Luna asks the hard questions about security, performance, and when this approach breaks down. If you've ever been tripped up by a migration that ran fine locally but failed in production, this episode is for you.

#WebAssembly #DatabaseMigrations #WASM #DeveloperTools #Startup #DevOps #CI/CD #Onboarding #Prisma #EdgeComputing #SQL #MigrationTooling #BusinessAndTechnology #FexingoBusiness #BusinessPodcast #TechnicalCoFounder #NoAds #Episode100

Keep every episode free: buymeacoffee.com/fexingo

In Episode 99, Lucas and Luna dive into OpenTelemetry, the open-source observability framework that is rapidly replacing proprietary APM tools like DataDog and New Relic. They explore how one startup—a fintech processing billions in transactions—switched from a costly APM vendor to OpenTelemetry to unify metrics, traces, and logs. Lucas explains the technical architecture: how OpenTelemetry's collector pipeline transforms telemetry data, reduces vendor lock-in, and cuts costs by 60 percent. Luna challenges the maturity concerns, asking about stability and missing features. They discuss the trade-offs with custom exporters and sampling strategies, and why engineers at this startup now prefer OpenTelemetry's flexibility despite the initial setup headache. By the end, listeners will understand why OpenTelemetry has become the de facto standard in cloud-native observability, and how it gives engineering teams control over their data without sacrificing performance.

#OpenTelemetry #Observability #APM #DataDog #NewRelic #Fintech #OpenSource #CloudNative #Metrics #Traces #Logs #Startup #Technology #Business #Engineering #FexingoBusiness #BusinessPodcast #TheTechnicalCoFounderPodcast

Keep every episode free: buymeacoffee.com/fexingo

In this episode of The Technical Co-Founder Podcast, Lucas and Luna explore how one startup leverages eBPF (extended Berkeley Packet Filter) to build a zero-latency database cache that runs inside the Linux kernel. They break down how eBPF allows the cache to intercept read requests at the system call level, bypassing the database entirely for hot data. The discussion covers the startup's architecture, the trade-offs of using kernel-level caching versus traditional Redis or Memcached, and how they achieve sub-microsecond latencies. Lucas explains the technical details: the eBPF program is attached to the `read` syscall, inspects the file descriptor, and redirects to a pre-populated hash map in kernel memory. Luna questions the development complexity and debuggability, and Lucas shares how the startup uses a fallback path to the database when cache misses occur. They also discuss the monitoring challenges and the startup's open-source tools for eBPF tracing. This episode is a deep dive into cutting-edge infrastructure for builders who want to push performance boundaries.

#eBPF #DatabaseCache #ZeroLatency #LinuxKernel #SyscallInterception #KernelModule #PerformanceEngineering #StartupArchitecture #Infrastructure #LowLatency #Caching #RedisAlternative #BPF #Technology #BusinessAndTechnology #FexingoBusiness #BusinessPodcast #TheTechnicalCoFounderPodcast

Keep every episode free: buymeacoffee.com/fexingo

Episode 97 of The Technical Co-Founder Podcast explores how a healthcare analytics startup called Synthos uses differential privacy to share sensitive patient data with researchers without violating HIPAA or GDPR. Lucas and Luna break down the epsilon parameter, the trade-off between privacy and accuracy, and how Synthos adds calibrated noise to aggregated queries to prevent re-identification attacks. They discuss a real incident where a hospital nearly leaked patient identities via a naive query system, and how differential privacy saved the partnership. The hosts also touch on practical implementation challenges like tuning epsilon per use case and the computational cost of privacy budgeting. If you're building products that handle sensitive user data, this episode offers a concrete framework for thinking about privacy engineering.

#DifferentialPrivacy #PrivacyEngineering #Synthos #HealthcareData #DataSecurity #HIPAA #GDPR #Epsilon #PrivacyBudget #NoiseCalibration #ReIdentificationAttack #Business #Technology #EngineeringFounders #CTO #FexingoBusiness #BusinessPodcast #TheTechnicalCoFounder

Keep every episode free: buymeacoffee.com/fexingo

Episode 96 of The Technical Co-Founder Podcast. Lucas and Luna explore how a logistics startup replaced a tangled mess of cron jobs and Bash scripts with a Directed Acyclic Graph (DAG) workflow engine built on Prefect. They break down the architecture: from dependency resolution and retry logic to real-time monitoring and cost savings. Discover why DAGs are becoming the standard for modern data pipelines and how this startup cut its ETL failure rate by 80 percent.

#DAG #Prefect #ETL #DataPipeline #WorkflowOrchestration #Cron #Logistics #Startup #Engineering #TechnicalCoFounder #Business #Technology #TechPodcast #EngineeringLead #DataEngineering #Python #FexingoBusiness #BusinessPodcast

Keep every episode free: buymeacoffee.com/fexingo