For data-intensive systems in Java, Kotlin, PostgreSQL — or all three. I diagnose the real bottleneck behind one specific performance problem, deliver a prioritized fix list, and walk your team through it. Fixed scope, fixed price, one week turnaround.
€6,000 fixed · 2–3 days of focused work · results within one week · read-only, nothing in your systems gets modified.
Who this is for
You have a specific performance problem: a slow endpoint, a query that stopped scaling, latency creeping up under load, infrastructure costs growing faster than traffic. Your engineers have profiled it, tried the obvious fixes, asked the AI assistants — and it’s still slow. That usually means the problem isn’t where the symptom is: it’s a layer up, in the architecture, and that’s exactly the kind of problem this audit exists for.
You don’t need a permanent hire. You need someone who finds the real cause fast and tells you exactly what to do about it.
What you get
- A written findings report — root-cause diagnosis and concrete recommended fixes, prioritized by impact vs. effort: what to change, why, and the expected gain.
- A live walkthrough call — I take your team through the findings, answer questions, and make sure every recommendation is actionable, not abstract.
The audit is diagnosis-focused by design: your engineers keep ownership of the codebase, and you get an expert’s map of where the problem actually lives — which is often not where the symptom appears.
Scope
- One focused problem per engagement — a slow query or endpoint, a scaling wall, a memory/GC issue, an architectural bottleneck.
- 2–3 days of focused work, results delivered within one week of start.
- Read-only. I don’t change your code or your data.
Price
€6,000, fixed. No day rates, no open clocks — you know the cost and the deliverable before we start.
Payment: 50% on booking, 50% on delivery of the report.
Day-one guarantee: if after the kickoff and a first look at your data I conclude this isn’t a problem I can crack, I’ll tell you on day one — and you pay nothing.
Why me
Senior software architect, 15+ years building and fixing data-intensive systems. Two areas where I go deep: the JVM — Java, Kotlin, Spring, concurrency, distributed systems, production AI workloads — and PostgreSQL, from query plans and schema design to what breaks under load in production.
When the head of one of the world’s top PostgreSQL consultancies was brought in to review database performance work I had done, his conclusion after a day on site: there was nothing he could add.
Recent write-ups of this kind of work: When Postgres is the wrong place to traverse the graph · Immutability vs concurrency: making a fast graph writable · Behavioral drift: silent bugs in LLM workflows
What I need from you
- A 60–90 min kickoff call with someone who knows the system.
- Production-grade diagnostic data. Real bottlenecks only show up in real data distributions and real load — staging numbers lie. You choose how I get there:
- Read-only access to monitoring/APM and database statistics (a read replica is fine), or
- Guided sessions — your engineer shares a screen and runs what I ask; I never touch your systems, or
- Exported artifacts — statistics snapshots, EXPLAIN (ANALYZE) output from production, metrics and log extracts.
- Read access to the relevant code — required for JVM systems: without it I can tell you what is slow, but not why. A read-only guest account (GitHub/GitLab) is enough; limited-module access or guided sessions work if full access isn’t possible. If the system isn’t on the JVM and the problem sits in the database, schema and query-level access is usually enough.
- Schema and architecture overview (docs or a whiteboard session — both work).
Everything is read-only. Nothing gets modified. NDA signed on request.
Optional follow-on
If you want me to implement the top fixes after the audit, we scope and price that separately. The audit stands on its own; implementation is your call.
How to start
Email me a two-paragraph description of the problem. I’ll tell you within 48 hours whether it fits the audit format — and if it doesn’t, I’ll say so and point you in a better direction.
Andrey Minogin · minogin.com · linkedin.com/in/minogin