Vladimir Lovtsov / Leadership notebook
Think deeper. Act with clarity.
On people, architecture, and the decisions we own. Practical analysis, professional observations, and questions I am working through.
Articles: 36
Analyst Days 14: Don't Paint the Grass — Data Risk in a BANI World
A reconstruction of my Analyst Days 14 talk on data quality and ownership: distinguishing a working control from attractive reporting.
Community Synergy: How a Large IT Company Can Use Its Own Expertise
How Tech Guilds helped me connect teams and knowledge. Reuse across a company, the structure of a Tech Hub, the economics of shared solutions and the conditions that make people want to participate.
Tech Lead: A New Challenge for an Analyst
Moving from analyst to project IT lead: building a team, establishing delivery practices, and discussing constraints with the business.
RAG for a company: from document search to a useful assistant
RAG explained from the ground up: indexing, retrieval, small-team and enterprise architectures, published implementations, evaluation, and the path from answers to useful actions.
AI in development: where it helps and where it creates new risk
A practical model-selection guide: 19 October 2026 API prices, self-hosting, task acceptance and the full cost of AI-assisted delivery.
Hiring Signals from a Manager’s Perspective
Behavioural patterns, an interview radar and the manager’s own blind spots: assess working risks without pretending to diagnose a personality.
The Shoemaker’s Own Shoes: Rebuilding My CV
How I removed technical noise from an executive CV, made the career trajectory visible, and established a clear CTO direction.
DPUPP on a Weekend: The Kremlin, Family, and an Engineer's Brain
A Sunday walk through central Moscow with my family, seen through the eyes of an engineer who still notices alerts and load tests away from the screen.
Building an Alternative Is Only the Start: Calculate the TCO
Compare a product, fork, compatible implementation, and bespoke build through total cost of ownership: upgrades, operations, skills, risk, and exit.
“Like FAANG,” Without the Cargo Cult: Bring the Context With the Practice
Architecture practices cannot be separated from scale, team structure, and operational maturity. Here is how to spot accidental complexity before adopting them.
Unhealthy Behaviour in a Tech Team: Signals and Actions
Observable signs of an unsafe working environment and practical actions for an individual contributor or manager, without diagnoses or labels.
EventStorming: Benefits, Limits, and Practice
A practical guide to EventStorming: what the method helps a team discover, where it falls short, and how to prepare a session that produces useful outcomes.
Are Retrospectives a Waste of Time? Sometimes
When a retro becomes a ritual, how to restore its purpose, and why a change without an owner rarely improves anything.
Taking a Banking Product Real-Time
Lessons from an early real-time product built in 2023–2024: streaming architecture, production testing, and delivery decisions.
Can an Overly Detailed Specification Weaken Engineering Initiative?
A working hypothesis about the relationship between requirement detail, delivery predictability, and team autonomy.
Five Common Systems Analysis Mistakes on IT Projects
Common systems analysis mistakes on large banking and fintech projects: vague requirements, legacy integrations, regulation, failure scenarios, and late validation.
Does an IT Professional Need a University Degree?
What university can genuinely offer an engineer or analyst, what that route costs, and when another form of education may be a better choice.
Data Exchange Formats: JSON, XML, Protobuf, Avro, and MessagePack
Choose a format through contract evolution, backward and forward compatibility, tooling, and measured processing cost rather than fashion.
Architecture Without Magic: Cost of Change and Operational Maturity
A decision guide for modular monoliths and microservices, based on change cost, ownership boundaries, and a team’s ability to operate a distributed system.
Building Fintech Products: Seven Paths to Production
A delivery map for banking products: team capability, governance, changing needs, infrastructure, cross-team contracts, the delivery platform, and operations.
UML in Practice: When a Diagram Helps and When It Does Not
Choosing a UML diagram for a specific question, defining how long it should live, and keeping modelling from becoming a project of its own.
Microservices or a Modular Monolith? A Decision Checklist
A concise decision guide based on domain boundaries, independent releases, scaling, data consistency, and the cost of operating a distributed system.
User Stories: From Need to Acceptance Criteria
How to frame a user story around who, what, and why, add testable acceptance criteria, and avoid turning it into a miniature specification.
Saint HighLoad++ 2024: Moving a Banking Product to Real-Time
How Vladimir Avramov and I turned an early real-time product built with Flink, Kafka, and Tarantool into a focused technical talk.
Word of the Day: Idempotency
What idempotency actually means in HTTP and distributed systems, why a UUID alone is not enough, and how to handle retries safely.
Building an IT Team Fast: Risks and Benefits
How to speed up hiring without pretending speed is free: what to test before the offer and how to reduce early hiring risk.
IT Conferences for Analysts: A Practical Guide
How to choose an event, prepare useful questions, and turn a talk or visit into durable professional connections.
How a Software Development Team Actually Works
One practical model for software delivery from discovery to release: six overlapping stages, common roles, and the team’s shared responsibility for outcomes.
SQL in Practice: Your First PostgreSQL Queries
A self-contained SQL exercise for analysts: create two tables and practise SELECT, filtering, JOINs, aggregates, data changes, and transactions.
Message Brokers: Where the Guarantees Stop
A practical comparison of RabbitMQ, ActiveMQ Classic, and Kafka: messaging models, acknowledgements, partition ordering, and the real limits of at-least-once and exactly-once.
A Career in IT: My Model for Choosing the Next Move
Four questions for comparing a management step, deeper expertise, or a move into a different kind of project.
API From A to Z: Semantics, Contracts, and Practice
How to choose between HTTP APIs, SOAP, GraphQL, JSON-RPC, and gRPC, then review methods, errors, compatibility, limits, and failure behaviour.
Analyst Roles in IT: A Map Without a Ranking
A practical, author-defined map of analysis roles based on the subject of the work, expected outcome, technical depth, and route into the field.
How to Interview a Client When You're the Analyst on an IT Team (a Checklist)
A step-by-step algorithm for interviewing a client when there is no business analyst on the team: preparation, running the interview, analysis, and confirming requirements.
NER for Business: From Entities to a Working Process
What NER actually does, how it differs from intent classification, and how to build a testable hybrid pipeline for Russian text.
Analyst Days 16: Behold the Path of Data, Padawan
A reconstruction of the main ideas from my Analyst Days 16 talk: how data gains context, why roles see the same dataset differently, and what an analyst should verify.
No articles found. Try another phrase or select all topics.
Short observations and new writing in my Telegram channel (in Russian).
@it_underside ↗