One IP, Six Crawler Identities, One Second: Detection Built Against Real Production Logs
How three production patches in 24 hours closed two leaks that synthetic testing missed
SpeyTech - Technical Articles
Technical articles on deterministic computing, safety-critical systems, and production AI architecture.
Latest posts
SEO Validator: A Deploy Gate for SEO Regressions
How an SEO audit that reads your nginx config can catch broken redirects before they reach production
Self-Hosted Static Sites Need Operational SEO Observability
How one open-source nginx log analyser surfaced both an acute crawler failure and a chronic crawl-rate suppression that conventional SEO tooling could not see
Undefined Behaviour: Compiler's Licence to Delete Your Code
How undefined behaviour allows C compilers to silently remove safety checks, reorder operations, and optimise away code that appears correct
Why memcmp Fails on Structs: Padding, Floats, Silent Bugs
How struct padding and IEEE 754 float representation cause memcmp to produce incorrect equality results, and why field-by-field comparison is the only fix
The Type Promotion Trap: C's Silent Integer Conversion Bugs
How implicit type promotion rules turn correct-looking comparisons into logic errors in C
Incident Reconstruction: Beyond It Worked Yesterday
How bit-perfect replay, execution tracing, and sealed audit logs transform incident response from guesswork to forensics
When Fixed-Point Beats Floating-Point (And When It Doesn't)
An honest analysis of when Q16.16 is the right choice and when floating-point is acceptable
Version Control for Deterministic Systems: Git Isn't Enough
How Merkle chains, cryptographic attestation, and reproducible builds satisfy certification evidence requirements
Init-Update-Status-Reset: O(1) Safety Guarantees
A four-function interface that enables static analysis, bounded resources, and compositional verification
Contracts as Documentation: Why Comments Lie
How preconditions, postconditions, and invariants become living documentation
Why 'Hello World' Fails Safety-Critical Engineers
Traditional C tutorials teach habits that certification auditors reject
C From Scratch: Learning Safety-Critical C
Why proving code correct before writing it changes everything
Testing ML Systems: Beyond Unit Tests and Accuracy Metrics
A practical testing strategy for production machine learning
Cost Engineering for ML Infrastructure
Where the money goes in ML infrastructure and what to optimise first
State Management in ML Services: Beyond Stateless Inference
Architectural patterns for ML systems that need to remember
Graceful Degradation in ML Systems
Fallback strategies for production inference that fails gracefully instead of failing loudly
The Observability Blind Spot: What ML Metrics Miss
Why accuracy looks fine while your production system burns
Fixed-Point Fundamentals
Learn fixed-point arithmetic from first principles — because 'close enough' isn't deterministic
certifiable-bench
Performance benchmarking for deterministic ML — because 'fast' means nothing if you can't prove it's correct
Stochastic Rounding Without the Stochastic
How PRNG-controlled rounding can provide regularisation benefits deterministically
Cross-Platform Bit-Identity: Theory to Practice
The practical journey of verifying deterministic ML across platforms
Feistel Shuffle: Deterministic Data Ordering
How cycle-walking Feistel networks can provide reproducible shuffling for ML training
Merkle Chains for ML Audit Trails
How cryptographic hash chains can make every training step verifiable
Round-to-Nearest-Even: The Key to Determinism
Why banker's rounding matters for bit-identical machine learning
The Certifiable-* Ecosystem: One Deterministic ML Pipeline
Eight interlocking C99 projects that make every ML pipeline stage bit-identical and auditable
certifiable-harness
End-to-end test harness for deterministic ML — because 'it works on my machine' isn't certifiable
certifiable-verify
Pipeline verification for the certifiable-* ecosystem — because 'we checked it manually' isn't certifiable
certifiable-monitor
Deterministic runtime monitoring — because 'the model drifted' isn't certifiable
Deterministic ML Pipeline for Safety-Critical Systems
How fixed-point arithmetic and cryptographic chaining produce bit-identical results from data to deployment
Certifiable Deploy
Deterministic model packaging and cryptographic attestation — because 'trust me, it's the right model' isn't certifiable
Certifiable Quant
Deterministic model quantization with formal error certificates for safety-critical ML
C-From-Scratch
Learn to build safety-critical systems in C — mathematical rigour, not 'Hello World'
Certifiable Data
Deterministic data pipelines for safety-critical ML — because 'we shuffled the data' isn't reproducible
C-Sentinel
Semantic observability for UNIX systems — lightweight system probing with explainable risk scoring
WCET Analysis for Neural Network Inference
How to prove worst-case execution time for convolution, matrix multiply, and pooling operations
Fixed-Point Neural Networks: The Math Behind Q16.16
How integer arithmetic can enable deterministic AI inference for safety-critical systems
TensorFlow Lite and DO-178C Certification Challenges
Understanding the architectural properties that complicate aerospace certification for mobile inference frameworks
Bit-Perfect Reproducibility: Why It Matters
What deterministic execution actually means and how to verify it across platforms
The Real Cost of Dynamic Memory in Safety-Critical Systems
Why malloc is problematic for certification and how static allocation can simplify verification
Certifiable Inference
Deterministic, bit-perfect neural network inference for safety-critical systems
Why Floating Point Is Dangerous for Deterministic AI
When 'mostly reproducible' isn't good enough for systems that matter
Certifiable Training
Deterministic ML training with Merkle audit trails — because 'we trained it' isn't certifiable
Debugging Model Behavior in Production
When the model works in staging but fails in prod, here's how to find out why
When You Don't Need a Feature Store
Most teams adopt a feature store before they have the problem it solves
Model Serving Architecture Patterns
Understanding latency, throughput, and the trade-offs between them
Production AI Systems: What 30 Years of UNIX Taught Me
The infrastructure principles that kept systems running still apply to ML
The Observability Gap in ML Systems
Why your model serving cluster fails at 3AM and you can't figure out why
From Proofs to Code: Mathematical Transcription in C
How mathematical contracts become deterministic implementations
The Hidden Cost of Non-Determinism
Understanding the financial impact of debugging race conditions and Heisenbugs