I am a Software Engineer (Backend — Go, TypeScript) with domain knowledge in FinTech, DeFi and e-commerce.

• TLDR: actively looking for a job now. 4 years YoE (mostly as Go/TS Backend in 2 BigTechs), 1+ years in DeFi at GotBit.io HF.

• Remote/hybrid/on-site: MENA 🇦🇪 , EU 🇪🇺 (+Cyprus 🇨🇾, +Turkey) ❤️ (I'm close to these TZs and have no problems with relocation). Also - 🇺🇸 & 🇧🇷, but I'm far away, maybe remote-only/relocation.

• Sharded a 1TB monolithic database into 24 shards without downtime.

• Acquired quite a strong expertise in professional optimization. Specifically, I've optimized gRPC handlers and Kafka consumers/producers. Either by low-level methods (pprof, metrics, etc.) or by refactoring architecture to eliminate bottlenecks.

• I created and designed algorithmic copytraing system for a Base Chain (chain id 8453).

• Deployed 2 blockchains from scratch - bootnodes, light-node and RPC EVM nodes on the base of AWS EC2 machines.

• I always loved to bypass antifrauds and begin career in tech based on that. However, my dream basically came true and I've created an antifraud from scratch for Avito - largest classifieds in the world. Specifically, I'm responsible for hardcore research, design review, and implementation of antifraud of ~50%+ payment traffic.

• Throughout the years, I've became pretty good at the so called DevSecOps - you can check out my linked in to see the work in this sphere. Honestly, I believe this is what every developer had to do.

However, besides job stuff I feel like I actually deeply love what I'm doing :)

• I kind of share the philosophy of Primigean - the more you do something, the better you get at it, the more interesting it is to do it. I was lucky I had 0 'boring' projects ever, maybe besides a few very first educational ones (coding decimal library in C11 decimal.h with bit-shifts was a bit boring, but actually I've used decimal type every day at crypto. I think this is Karma 😁)

• TJ, especially his PDE Talk and tutorials ~3 years ago inspired me to give NeoVim a chance. Since that I'm a happy user of Neovim. I definitely can appreciate other editors, but NeoVim is just so good/extensive (and I also like what the team is doing with architecture, RPC protocol, etc.), that I see no reason to change.

• POV: I'm on my third coffee and you just asked me how the internet works

• I have gave Nix & NixOS a fair change for 3 weeks, but I'm not sure they actually solve problems they actually solve problems they claim to solve. (maybe I missed proper intro by @alurm, who knows..)

• My Journey to Saga (2025/01/13)

  • About: this blogpost is my journey to how I finally understood Saga microservices design pattern with 0 magic 🙂 Quite funny, that I have seen it or implemented myself throughout the years, but just didn't know the fancy name. Now I've written a short article to help other engineers figure it out.
  • Prerequisites: know basics of microservices arch, generally be familiar with idempotency, ACID in RDBMS, network IO & related issues that follow network.
  • Read time: ~10 minutes
  • Why Read: It would be awesome if you didn't understand the pattern before, but this text helped you grok it 🎉 I have actually worked with them since day 1 of my career, but no one in team called it like that. I think this shows that the concept it quite powerful. But definitely it requires respect to software complexity, it may be easy to over-use them.

• Discord Antifraud UX — Case Study

  • About: a small case study about the discord antifraud UX
  • Prerequisites: none
  • Read time: ~2 minutes
  • Output: this article gives basic info what is antifraud, my opinion on the discord UX. Also there is a chance you will fall in love intro reversing these things as I did 😅

• tbd: why FZF-go should be forked (sorry, I use fzf daily and love it, the algo is great, but the CLI experience ...)

  • terrible go code, not the core algo, but CLI experience - just terrible
  • terrible code => panics everywhere => fails silently => hard to debug
  • terrible devex with super long --abcdd options
  • not an easy to use programmatic go pkg API. Everything in one package - oh no 🙄
  • go code is so terrible & outdated, that complete rewrite would be better
  • I see several minor optimizations by just making code slightly better
  • making --abcd flags into actual API and separate binaries for users (from my estimation, no benchmarks yet) would dramatically improve performance in some cases
  • http api is .. 💀 not that good.

  why fork this project and not fzy (c) or skim (rust) ? Because of skill issue of course. Actually, it's easier to fork it + it has really good algorithm that I see no reason to change - it's just the wrapper is really outdated..

• tbd: hype that 'devs are no longer needed

  • doesn't match my experience as a reasonably skilled mid+/senior engineer (not about code, but about

Feel free to reach me out with any questions, feedback or just to say hi. Errata is also appreciated 🙏

Feel free to talk to me in:

— English (C1) - my eng. speaking friends call me Mike — Russian (native) - my Russian name is Misha (Mikhail)

Using these contacts (telegram is strongly preferred):

• Telegram - @mikhailnkuz
• My Telegram Channel - @mikhailnkuz_channel
• LinkedIn - /in/mikenk/
• Mail: [email protected]
• 42 Alumni - @posidoni (@posidoni is just a random handle I've got at 42 and quite liked)

Notice: all opinions are my own and not the views of my employer/contract counterparty (past or present) or any other entity I'm affiliated with. Do your own research.