Technical Writing
Tech Articles
Practical implementation notes from client work: AI agents, RAG pipelines, agentic coding, AEO, GEO, and production AI systems.
Medium archive: https://medium.com/@max.petrusenko
Featured On-Site
Canonical tech guides published here.
Bitcoin as Strong MoneyWhy Bitcoin is treated differently from generic crypto: scarcity, self-custody, neutral settlement, and the tradeoffs that still matter.
Answer Engine Optimization (AEO) GuideHow to structure content so AI answer engines can discover, extract, and cite your expertise.
OpenClaw Installation Playbook for TeamsDeployment checklist, security guardrails, and rollout sequence for OpenClaw installations in real client environments.
All Tech Topics
24 articles across AI, RAG, agents, and more.AI Infrastructure for Startups: Minimum StackThe minimum production setup for AI products: model gateway, retrieval layer, observability, and evaluation.
LLM Serving Architecture: Latency and Cost ControlsArchitecture patterns to reduce response time and token spend without sacrificing output quality.
AI Infrastructure Security BaselineCore controls for secrets, access boundaries, auditability, and incident response in AI systems.
LLM Evaluation Metrics That Actually MatterA concise framework for selecting evaluation metrics that map to business outcomes and reliability targets.
Build an LLM Eval Dataset from Production TracesHow to convert real user interactions into reusable test sets for regression and model comparison.
Offline vs Online LLM EvalsWhen to run synthetic benchmarks, when to measure in production, and how to combine both.
LLM-as-Judge Rubric DesignRubric patterns that improve consistency and reduce evaluator drift in LLM-as-judge pipelines.
Pairwise vs Absolute LLM ScoringTradeoffs between pairwise comparisons and absolute scorecards for prompt and model selection.
Tool-Calling Evals: Schema and RetriesEvaluate function-calling reliability with schema compliance, retries, and side-effect safety checks.
Agent Evals: Trajectory QualityHow to score multi-step agent behavior, tool choice, and completion efficiency.
CI/CD Eval Gates for LLM AppsA release pipeline pattern that blocks regressions with automated eval checks.
Hallucination Testing: Reference-Based and Reference-FreeTesting methods to catch factual drift, unsupported claims, and citation mismatches.
LLM Regression Dashboard: Alerts and ThresholdsDashboard design for evaluation regressions, quality alerts, and deployment control loops.
Hybrid Search 101: BM25, Vectors, and RerankingA practical baseline for combining lexical and semantic retrieval with rerankers.
RRF vs Weighted Fusion for Hybrid RankingHow reciprocal rank fusion compares with weighted scoring for hybrid retrieval systems.
Hybrid Retrieval with ACL and Metadata FiltersDesigning hybrid retrieval that respects tenant boundaries and metadata constraints.
Hybrid Search in Elasticsearch: Practical PatternsQuery patterns and tuning workflow for hybrid retrieval in Elasticsearch deployments.
Hybrid Search in Weaviate: Alpha TuningHow to tune alpha and query parameters for balanced lexical-semantic retrieval in Weaviate.
Hybrid Search in Qdrant: Quality MeasurementImplementation notes for Qdrant hybrid queries with relevance measurement and diagnostics.
Query Routing: Lexical-First vs Semantic-FirstRouting strategies for deciding when lexical retrieval or semantic retrieval should lead.
Hybrid + Reranker Architecture for Support AssistantsA retrieval architecture for support bots that balances recall, accuracy, and response speed.
Hybrid Retrieval for Long-Tail and Exact-Match QueriesHow hybrid stacks handle precise terminology and sparse long-tail intents without quality collapse.
Hybrid Retrieval Debugging: Why Irrelevant Chunks WinA debugging workflow for noisy retrieval results in hybrid pipelines.
RAG Pipeline Architecture End-to-EndEnd-to-end blueprint for ingest, indexing, retrieval, generation, and evaluation in RAG systems.