mikenkuz

Senior Golang Engineer

Senior Backend Engineer (Go) · 5 years of experience · ex. Ozon, Avito, inDrive · Strong FinTech (PCI DSS) domain experience · Designed and delivered high availability systems

· 𝐎𝐳𝐨𝐧 (e-commerce, sellers): Designed a plan and sharded a 1TB PostgreSQL monolith into 24 shards with zero downtime in production. Sped up Kafka pipelines 2.5x. Implemented complex product projects that influence millions of sellers and items daily. Result: sellers' items are available for purchase near instantly after item creation.

· 𝐀𝐯𝐢𝐭𝐨 (FinTech, payments): Built the antifraud solution for SBP (Fast Payment System), protecting 50%+ of all Avito payment traffic. Passed PCI DSS certification and committed directly to services that handle credit card payments. Communicated directly with the payment provider to troubleshoot a defect.

· 𝐢𝐧𝐃𝐫𝐢𝐯𝐞 (ride-hailing, pricing): Owned the real time pricing backend across 14 AWS EKS clusters and rebuilt the service's MCP layer for full production observability. Also built an SRE agentic AI solution that produced insights from raw signals (logs, traces, Kubernetes cluster info, etc.). Coordinated and troubleshot 2 incidents affecting the recommended price feature.

· 𝐁𝐥𝐮𝐦 (Telegram app, crypto): Built a Telegram Mini App (@Blum) alongside my teammates. Moved it from a JavaScript MVP to Go, profiled it to handle 200K concurrent users and 22M daily active users, and built the Telegram Stars payment gateway.

· 𝐆𝐨𝐭𝐁𝐢𝐭 (crypto trading): Shipped 2 production EVM chain forks. Built Solana and TON backend integrations. Handled client communication. Architected and delivered a copy-trading system on Base Chain (EVM Layer 2).

𝐒𝐭𝐚𝐜𝐤
Go, PostgreSQL, Kafka, Redis, AWS, GNU/Linux, Kubernetes (Helm, EKS, ArgoCD), HTTP, gRPC, MCP, plus Web3 (EVM/geth, TON), ClickHouse, Databricks, K6, pprof, GraphQL, TypeScript, Rust, GitHub Actions, GitLab CI/CD

𝐎𝐩𝐞𝐧 𝐭𝐨
· Senior Backend Engineer roles
· DevOps/SRE and similar infrastructure positions
· Agentic AI engineering jobs

𝐂𝐨𝐧𝐭𝐚𝐜𝐭𝐬
· Email: [email protected]
· Telegram: t.me/mikhailnkuz (@mikhailnkuz)
· GitHub: github.com/posidoni

𝐋𝐚𝐧𝐠𝐮𝐚𝐠𝐞𝐬
English - fluent, Russian - native

𝐋𝐨𝐜𝐚𝐭𝐢𝐨𝐧𝐬
· Current: Tbilisi, Georgia, UTC+04:00
· Open to full remote or on-site/hybrid (EU, EMEA, Gulf). Georgia, Cyprus: high priority
· Former Cyprus Key Employee residence permit holder, all paperwork ready

Experience: 5 years

Yearly salary: $60,000

Hourly rate: $0

Nationality: 🌏 Remote

Residency: 🇬🇪 Georgia

Experience

Senior Software Engineer

Blum

2025 - 2025

Blum is a Telegram-based crypto exchange founded by ex-Binance executives. I was a senior engineer on @Blum, the main bot and Telegram Mini App launcher. At peak it served 200K concurrent users and 22M daily active users. Reason for leaving: the startup hit economic hardships that led to mass layoffs, including all second-level managers and the key product owner. · Migrated @Blum from the legacy JavaScript codebase to Go. I profiled hot paths with pprof and made sure the new architecture would handle the load. This cut p99 latency and memory consumption, and unblocked the team to ship new features. · Built the Telegram Stars payment flow in the bot. In the first 2 months it earned 300K+ Stars (~$4K) and unlocked Telegram's mass-message Bot API, resolving a long-standing rate-limit problem. · Took the SRE initiative after seeing issues during the first stage of the TGE (Token Generation Event): resolved and coordinated incidents, built Grafana dashboards, introduced an SDK for metrics, logging, health checks and rate limits, and reduced cloud provider bills. During the final critical stage of the TGE we had zero critical issues.

Senior Software Engineer

inDrive

2025 - 2026

inDrive is a global ride-hailing platform. On the Pricing Team I owned the recommended price - every user sees this number when requesting a ride. · Recommended price: Owned the Go services computing the ride price in real time across 14 AWS EKS clusters, with geo-pricing on Uber's H3 hexagonal grid, plus the tariff back office used to manage pricing configuration. · Go platform contribution: Fixed RESP3 negotiation in the company's Redis client library. This removed two redundant round trips per connection for 100+ Go services. Merged after review by two principal engineers. · SRE bot: Built an LLM agent (Claude API) that pulled metrics, logs, traces and Kubernetes state, cross-referenced them with code and Slack, Confluence & Jira history (Grafana, GitHub, Slack, Atlassian and PagerDuty MCPs). In order to optimize context usage, datasources were analyzed by sub-agents; troubleshooting workflows were codified as reusable skills. During one live incident it traced a pricing misconfiguration to a 2-year-old ticket and linked it to an A/B experiment. This allowed us to quickly mitigate the issue by turning off the A/B toggle. · On-call & SRE: Weekly on-call rotation (PagerDuty). Coordinated and troubleshot 2 production incidents and wrote postmortems. Added business-critical alerts to Alertmanager and tuned legacy ones, tracked SLA/SLO, wrote runbooks, documented safe deploy procedures. Stack: Go, Python, AWS, Kubernetes (EKS), GitHub Actions, ArgoCD, Temporal, MySQL, Redis, Jaeger, Grafana, Thanos, gRPC, MCP, Vault

Senior Software Engineer

Avito

2023 - 2025

Avito is the world's largest classifieds platform, with over 80 million MAU. It enables buying, selling and renting across many categories. · I authored antifraud for 50%+ of payment traffic in Avito with the SBP (Fast Payment System) method. I did the research, proposed the design plan and got it approved by TLs and senior engineers in the Payments domain. Then I drove the project as a feature leader, overseeing and implementing tasks myself. Result: the majority of SBP traffic is protected with best-practice antifraud. · I diagnosed and resolved a complex bug in the Credit Scoring service: a PostgreSQL + PgBouncer connection was unexpectedly exchanged mid-transaction, which broke score recalculation. Reliability improved from 75% to the expected 99.99%. The service now calculates credit scores for 2M+ users daily without database issues. To prevent regressions, I added logs, metrics, traces and alerts. · I proposed and completed a redesign of the antifraud engine to a schemaless data model. The engine became more flexible for API partners. Dev productivity and test coverage also grew: all components became generic by default (Go 1.22).

Software Engineer

GotBit HF

2022 - 2023

Gotbit was an International Crypto Hedge Fund, one of the largest Liquidity Providers and Market Making companies. I was a part of VC team and I along other engineers I was responsible for building products both for a GotBit and its clients. • I developed algorithmic copytraing system for a Base Chain (chain id 8453). During this project I have researched best routing API and organized calls with Head of API of Li.Fi and Zerion team. Eventually, Zerion was used for this project as paid routing API provider. • Created and supported an internal microservices API for blockchain events data scanning (logs, tx.data), offering PubSub and Polling API. This solution has significantly reduced costs for a Node RPC providers, and simplified our internal usage. • Implemented SDK adapters for top Ton DEXs (StonFi, DeDust). I reverse-engineered DeDust API, discovered a private routing API and got a permission from founder to use it. This significantly accelerated development and we have successfully integrated these adapters for a GotBit trading platform. Result: GotBit trading platform has been successfully used on Ton to provide MM services.

Software Engineer

Ozon

2020 - 2022

Ozon is one of the leading CIS e-commerce companies: ~5k engineers, 1B+ product cards. · PostgreSQL sharding: I planned and executed the split of a 1TB monolithic database into 24 shards with zero downtime in production. This significantly improved the reliability and correctness of the most loaded service in our domain. · Instant item availability: I designed and built a new service with a gRPC API and REST (HTTP/JSON) handlers for the frontend. Result: sellers' items became available for purchase almost instantly after creation. This affected every seller on the platform. · Kafka: I optimized consumers and producers under strict resource quotas and made our queues up to 2.5x faster. I collected metrics with Prometheus, Grafana and Thanos, set up alerts and profiled with pprof. Along the way I learned Kafka in depth: heartbeats, leader election, message keys, topic partitioning. · Measurement automation: I designed and developed a service that automates 500K+ product measurements daily. It matches similar items via an ML service over gRPC: given 20 similar iPhone cases, a warehouse employee now measures only 1. Strict performance requirements were met with a simple PostgreSQL queue and cheap Redis locks.

Skills

api

aws

backend

ethereum

evm

fintech

gcp

golang

javascript

postgres

typescript

english

russian

Create Profile Hire Golang Developers