Articles
About the author
I'm Roman, a Lead AI/ML Engineer with 15+ years of software development. I take AI products from idea to production: LLM agents, RAG, text-to-SQL and MCP servers, with the backend, data, integrations, cloud and security behind them. I also do a lot of R&D and project estimation, work with clients as a Forward-Deployed AI Engineer, set engineering standards and mentor developers. Before that I built web, backend and mobile applications for banks, government and business.
GEO, MCP and WebMCP: how ChatGPT, Claude and Gemini find, read and use websites
GEO in practice: how AI assistants find a site, what to set up in robots.txt, markup and content, how to connect MCP and WebMCP, and what an online store needs for AI-driven purchases.
Python for Java Developers
Python for those who know Java: syntax and core methods next to Java, strings, collections, generators, threads, asyncio and a microservice on the standard library.
LLM inference engines: vLLM, llama.cpp, LiteRT-LM and others — speed, hardware, scaling
What to run language models with in late 2026: the engines vLLM, SGLang, TensorRT-LLM, llama.cpp, MLX, LiteRT-LM and ExecuTorch, and programs built on them such as Ollama and LM Studio — what makes the engines fast, what hardware they run on and how they scale, with measurements on a laptop.
Comparison of 41 AI harnesses — from Claude Code to OpenClaw
Coding agents and general-purpose agents compared: Claude Code, Codex, Cursor, Copilot, OpenCode, OpenClaw, Hermes — licenses, models, configuration, sandboxing, benchmarks, training on your code and leaks.
Claude Code — commands, settings, flags
A Claude Code reference: every slash command with its arguments broken down and examples, settings.json keys, permission rules, terminal subcommands, flags and environment variables. Current as of August 2026.
Claude Code — Best Practices
How to organize work with Claude Code: project instructions, rules and skills, a task and bug tracker in the repository, planning before code, subagents, context, permissions and budget.
Claude Code Hooks
A walkthrough of Claude Code hooks: every event, exit codes and JSON decisions, matcher and if, the five handler types, and three working hooks — sound, subscription limits and a work log.
LangChain and LangGraph — agents, graphs and the infrastructure around them
How the LangChain 1.x and LangGraph 1.x stack fits together: models and messages, tools, create_agent and middleware, StateGraph with reducers, checkpointers and store, streaming, human-in-the-loop, LangSmith, Agent Server and the langgraph CLI, plus the 0.x migration.
Prompt engineering techniques for Large Language Models
The effort knob on reasoning models, strict output schemas, agent prompts, delimiters, XML tags and few-shot — how to write prompts for ChatGPT, Claude or Gemini and get predictable output.
DSPy — Automatic Prompt Generation
DSPy 3.4 through a support agent: signatures instead of prompt text, tools, a metric, the MIPROv2 and GEPA optimizers, saving the result, and what changes in DSPy 3.5.
RAG (Retrieval-Augmented Generation), Function / Tools Calling, CAG (Cache-Augmented Generation) for LLM models
RAG, function calling and CAG for local LLMs: how vector search works, with working examples in Kotlin, Dspy, LangChain, LangChain4j and llama.cpp.
Vector databases
How vector databases work: embeddings, HNSW indexing and similarity search, plus a comparison of Milvus, Qdrant and Weaviate with Milvus code.
Offline AI Launcher — launch AI LLM neural networks on smartphone
Offline AI Launcher runs LLMs like Gemma, DeepSeek, Llama and Qwen straight on your Android phone from .gguf files, with no internet or subscription.
LoRA — Fine-Tuning AI LLM Models
Fine-tuning an LLM with LoRA: when it beats RAG and when it does not, how the adapters work, and two complete examples for Gemma 4 E2B on Unsloth and PEFT with export to GGUF.
Kubernetes — How it works
How Kubernetes works: control plane architecture, namespaces, load balancing, everyday kubectl commands, the Dashboard and Prometheus monitoring.
gRPC — How it works
How gRPC works on top of HTTP/2 and Protocol Buffers, where it beats REST for internal services, and how to build a Kotlin server and client.
llama.cpp AI LLM Engine — How It Works
How inference works inside llama.cpp: loading a GGUF model, tokenization, building the compute graph, the KV-cache, and token sampling.
Running LLMs locally on an Android smartphone
What can run a language model right on Android: LiteRT-LM, llama.cpp, MediaPipe, MLC LLM and ONNX Runtime GenAI — which to choose, how to prepare a model, and Kotlin code that stops generation correctly.
Neural networks in simple terms
How an LLM actually works: tokenization, embeddings, transformer attention, quantization, and the samplers that pick each next token.
Introduction to Project Loom
Project Loom virtual threads in Java: how the JVM schedules them on carrier threads, how they differ from Kotlin Coroutines, and enabling them in Spring.
Introduction to Ktor and Exposed
Building a Kotlin server with Ktor: Exposed and JDBI for the database, Koin for dependency injection, Mockk and Testcontainers tests, Flyway migrations.
Kotlin 2.0 — Key Syntax Changes
What changed in Kotlin 2.0 syntax: context receivers, value classes, sealed interfaces, destructuring in for loops and pattern matching in when.
Relational SQL databases
SQL syntax, views and indexes, JOIN and CTE queries, pessimistic and optimistic locking, and reading query plans in PostgreSQL, MySQL and Oracle.
Microservices and Web Application Security
Auth microservices, API Gateway, mTLS and certificates: how to store passwords safely, stop brute-force attacks and choose encryption algorithms.
Java Spring — Cloud
Cloud services for Java across AWS, Google Cloud and Azure: side-by-side equivalents, cost optimization, serverless, Git LFS and submodules.
Java Spring — Spring Components
Configuring application.properties, WebSocket, GraphQL, OAuth 2.0, Spring Security, Actuator, WebFlux and Batch with real code and config examples.
Java Spring — Concepts
Core backend concepts behind Spring: Saga and Rollback, SOLID, Clean Architecture, REST vs RPC vs GraphQL, IoC and DI, Spring MVC and Project Reactor.
Java Spring — Data Storage
Data storage in Spring: ACID and isolation levels, @Transactional, JPA vs JDBC, Spring Data repositories, Redis caching, Flyway and indexes.
Java Spring — Microservices
Microservice communication in Spring: REST and WebClient, Kafka and RabbitMQ, Docker and Kubernetes, Eureka, Cloud Gateway, and request tracing.
Java Spring — Spring Annotations
A reference to Spring annotations with Kotlin examples: @Component, @Bean, @Transactional, request mapping, JPA, and how @Component differs from @Bean.
Java Core — Basic Questions
Java Core interview answers: data types, collections and HashMap, generics, reflection, JVM memory layout and how the garbage collector works.
Java Spring — List of Questions
A navigable index of Java developer interview questions: Java Core, Kotlin, concurrency, Spring annotations and components, microservices, and SQL.
Setup custom launch options for IntelliJ IDEA / Android Studio
What each flag in Edit Custom VM Options actually does: heap sizes, Garbage Collector choice, code cache, and memory tuning for Android Studio.
Small manual on MapStruct
Practical MapStruct recipes for Kotlin: mapping mismatched field names, @Named methods, reusing mapping logic, and avoiding common compile-time errors.
Google Guava in Examples
A tour of Google Guava through Kotlin examples: Multimap, BiMap, Table, caching, string utilities, hashing, IO helpers and concurrency.
Logging Android application work in txt files
A LogUtil logger for Android that writes debug and error logs plus OkHttp network traffic into separate txt files on the device storage.
Dynamic UI Generation in Android Application Using RecyclerView and AdapterDelegates
Building dynamic Android UI with RecyclerView and AdapterDelegates: writing delegates, reusing list items, and diffing them without DiffUtil boilerplate.
How to set up Charles Proxy for the Android Studio emulator
Step-by-step Charles Proxy setup for the Android emulator: proxy via adb, the Charles certificate, a network security config for your app's HTTPS, and rewriting responses.
Mock Server Responses in Android Development using Interceptor
Mocking server responses in Android with a custom OkHttp Interceptor: a MockResponse model, a URL-keyed map, and wiring it into OkHttpClient.
Android — Components and Views
Android internals for interviews: Activity, Fragment and Service lifecycles, Context, Intent, WorkManager, RecyclerView and MVP vs MVVM vs MVI.
Android Paging 3 and Paging in AdapterDelegates
Building pagination in Android with Paging 3: PagingSource, PagingDataAdapter, load-state footers, and wiring it into AdapterDelegates.
Android Push Provisioning API
Adding bank cards to Google Pay from an Android app: tapAndPay SDK setup, API allowlisting, Google Pay UI rules, and encrypted OPC handling.
How to enable Wi-Fi debugging in Android Studio
Debugging an Android app over Wi-Fi: the classic adb tcpip and adb connect route, and Android 11+ wireless debugging with pairing by code or QR.
DexClassLoader, or loading part of an app from the network
How an Android app downloads a .dex file from a server and loads classes from it with DexClassLoader, what Android 14 restricts, and why Google Play forbids it.
Java / Kotlin Core — Multithreading
Java and Kotlin concurrency explained: volatile, synchronized, ReentrantLock, coroutines and Channels, RxJava operators and backpressure strategies.
Kotlin Core — Basic Questions
How coroutines and suspend actually work, when to use launch instead of async, plus scope functions, delegates, extensions and companion objects.
Java / Kotlin Core — Theoretical Questions
Big-O costs of collection operations, design patterns, SOLID, inheritance versus composition and higher-order functions in Java and Kotlin.