Skip to content
View alex-feel's full-sized avatar
😎
Focusing on what's important right now
😎
Focusing on what's important right now
12 followers · 3 following

Achievements

Achievement: Pair Extraordinairex2Achievement: StarstruckAchievement: YOLOAchievement: Pull Sharkx3Achievement: Quickdraw

Achievements

Achievement: Pair Extraordinairex2Achievement: StarstruckAchievement: YOLOAchievement: Pull Sharkx3Achievement: Quickdraw

Block or report alex-feel

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
alex-feel/README.md

Aleksandr Filippov

Lead Product Manager, AI Platforms & Agents — I turn frontier AI capabilities into reliable products and reusable platforms.

I build applied AI systems end to end: agent platforms, enterprise retrieval, MCP servers and gateways, and the developer tooling around them. Product ownership and production Python in the same pair of hands — problem discovery, quality criteria, architecture, implementation, deployment, adoption.

How I build

  1. Product ownership — I start with the problem: users, constraints, success criteria, and the adoption path — before any code.
  2. Platform thinking — I build reusable capabilities — retrieval, gateways, agent runtimes — rather than isolated AI demos.
  3. AI-native engineering — Coding agents multiply my implementation speed; architecture, review, and accountability for outcomes stay mine.
  4. Evaluation and verification — I define invariants, regression tests, and quality gates before trusting model or system behavior.
  5. Production and adoption — A system counts when it is deployed, observed, supported, and measurably used.

Selected work

  • MCP Context Server — persistent agent memory: full-text, semantic, and hybrid retrieval behind MCP, on PyPI and in the official MCP Registry
  • Claude Code Toolbox — one-command Claude Code environments for real teams
  • CXR Draft Auditor — a fine-tuned 4B medical model paired with a stock 4B parser, with a held-out evaluation harness
  • The enterprise systems — a shared retrieval platform, a production YouTrack MCP server, an AI localization product with measured savings — are written up on alexfeel.info

Elsewhere

alexfeel.info · LinkedIn · X

Pinned Loading

  1. mcp-context-server mcp-context-server Public

    MCP Context Server — a FastMCP-based server providing persistent multimodal context storage for LLM agents.

    Python 9 4

  2. claude-code-toolbox claude-code-toolbox Public

    Claude Code Toolbox — automated installers and environment configuration framework for Claude Code with one-line setup across Windows, macOS, and Linux.

    Python 16 4

  3. cxr-draft-auditor cxr-draft-auditor Public

    Research/educational chest X-ray draft auditor: image-grounded findings vs draft impression, with a deterministic comparator. NOT a medical device.

    Python 1

  4. claude-code-artifacts-public claude-code-artifacts-public Public

    Reusable Claude Code artifacts — skills, hooks, rules, environment configs, agents, and slash commands — ready to drop into any setup.

    Python 6

  5. st-copy st-copy Public

    A tiny, theme‑aware Streamlit component that adds a one‑click "copy-to-clipboard" button to your app — perfect for the chat UI, URLs or any other text the user might need to copy.

    Python 9 1

  6. hugo-artifacts hugo-artifacts Public

    Reusable, style-agnostic Hugo modules — shortcodes, asset & utility modules that ship semantic BEM markup and zero CSS, so they drop into any number of sites and you style them to fit.

    HTML 1