Banner Image

Skills

  • TypeScript
  • Algorithms
  • Analytics
  • Bitcoin
  • Bitcoin Development
  • Computer Science
  • Cryptocurrency
  • Cryptography
  • Functional Programming
  • Game Development
  • Haskell
  • Mathematics
  • Minimum Viable Product (MVP)
  • Mobile Development
  • Python

Sign up or Log in to see more.

Services

  • MVP Development, Research

    $50/hr Starting at $2K Ongoing

    Dedicated Resource

    Experienced software engineer - over a decade. Product-oriented mindset. US/EU timezone.  I started as Java/Scala/Haskell developer - currently develop full-stack, whole solutions. Domains: Quantitative...

    AlgorithmsAnalyticsBitcoinBitcoin DevelopmentComputer Science

About

Product Engineer

Over a decade of experience. Theory and practice. Started with Java/Scala, then Python/Haskell, then Typescript. Can work with any other language or technology.

Developed external DSL and various solutions for a major bank. Developed Natural Language Processing, worked with theorem proving and lambdas. Developed for Bitcoin/Cardano, published a Financial Cryptology paper in Springer's LNCS. Developed MVP of proof-of-work based fact certification with Wolfram/Cardano.
Developed my own mobile game.




Know cryptography (algorithms, mod-inverses) and mathematical abstractions behind it.


Hobbies.

I travel a lot. I know that true market price on food is unaffordable. I know that pseudo-sciences and beliefs in "infinite objects" (used to "prove" Turing-Church thesis among other things) are result of agricultural decline. I know that halting problem is actually decidable, not undecidable semi-decidable. In its standard formulation: for all inputs, inc dec jmpnz abstract machine.

I know that asymmetric encryption is "cat in a bag" security - defender has to spend as much energy/resources as attacker to ensure security of a given pubkey.
Statistical estimates of "overheads of fast exponentiation for DLP", which NSA and others (eg Blackberry) are using as excuse are not applicable to non-monotonic problems, foolish useless soviet 'proofs' there, coming from worst european ideas.

Additionally, ECC is even worse than DH, speaking of crypto - polynomials introduce more holes on top of illusions.

I proved it all to GPT-OSS, only to realize how limited it is, since I could also prove exact opposite to it. And OpenAI "improves" their models by stealing mine and others' ideas from private chats (duck.ai and cloud services including).
I think transformation grammars would work better on same dataset, it agrees.

I proposed better data source (search engine) for AI where PoW is used to certify facts and "reverse word/pattern" relevance of results (see YaQui search on my Upwork). BTC for support: bc1qekvmkczge3hxrvwdf2lj3yyvgjnparn3fdf9lg.

P.S. Note about math proofs: actual working code, not sage math, not even proof assistant (wrong assumptions, unsound inappropriate abstractions).
The code (grounded in reality, with 100% code overage) is the only reliable proof. I have some on github.

Work Terms

I need clear description of real world problem or goal. I can design requirements and architecture for you. I don't work in teams, unless it is actually needed which is extremely rare. I accept bitcoin, SEPA, SWIFT. I have legal entity in Ukraine for B2B. I have residency permit in EU.