I design and ship production systems at the intersection of software engineering, machine learning, and MLOps—from data pipelines and model deployment to APIs, automation, and applied research. I work end-to-end with product and platform teams to keep systems reliable, observable, and scalable.

• Production ML & agents: LLM/RAG workflows, fine-tuning, coding-agent and SWE-bench–style tooling, evaluation and guardrails in real deployments.
• Data & platforms: ETL, streaming-style pipelines, monitoring, and infra patterns (containers, cloud, Terraform) that support ML and product workloads.
• Web & integrations: Full-stack and API work (Node, React, Django/Flask), Discord/Telegram automation, VS Code extensions, and enterprise platforms where relevant.
• Applied research → product: Computer vision, NLP, classical ML, and notebooks that graduate into maintainable services—not one-off experiments.

Consulting-style work (where applicable): game automation, community/token distribution mechanics with anti-abuse, trading and market data tooling, Web3 (TON/Tact, smart contracts, NFT backends).

Stack and tooling below reflect public and private work and the languages surfaced from collected GitHub metadata.

Languages: JavaScript / TypeScript, C / C++, AMPL, Astro, Batchfile, Blade, C#, CMake, CoffeeScript, CSS, Cuda, Dockerfile, EJS, Elixir, Go, Go Template, Handlebars, HTML, Inno Setup, Java, Jupyter (notebooks), Kotlin, Less, Makefile, Mako, MATLAB, mcfunction, MDX, MQL5, Mustache, Objective-C, Objective-C++, PHP, PLpgSQL, PowerShell, Python, R, Ruby, Rust, SCSS, Shell, Solidity, Starlark, Swift, Tact (TON), TeX, Vue, XSLT, SQL

ML & AI: Classical ML and deep learning, computer vision, NLP, PyTorch, TensorFlow, Keras, OpenCV, MediaPipe, EfficientNet, StyleGAN2, neural ODEs, recommendation (e.g. LightFM), LLMs, RAG, LangChain / LangGraph, QLoRA / fine-tuning, coding agents and SWE-bench–style evaluation.

Data & analytics: ETL, visualization, time series, statistics, data quality, and modeling for decision systems.

Web & applications: Node.js, React, Django, Flask, Discord bots, VS Code extensions, ServiceNow flows, Drupal.

Cloud, infra & MLOps: Azure, Terraform, container workflows, Cloudflare R2, active-container style deployments.

Other domains: Blockchain and Web3 (IOTA, TON, Bittensor agents), exchange and market data, quantum demos (QFT), trading and strategy analysis (non-personal).