Skip to content
View DurdeuVlad's full-sized avatar
5 followers · 1 following

Achievements

Achievement: Pair Extraordinairex3Achievement: Pull Sharkx3Achievement: QuickdrawAchievement: YOLO

Achievements

Achievement: Pair Extraordinairex3Achievement: Pull Sharkx3Achievement: QuickdrawAchievement: YOLO

Highlights

  • Pro

Block or report DurdeuVlad

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
DurdeuVlad/README.md

Vlad Durdeu

Python & AI engineer building reliable backend systems, agentic workflows, and developer tools.

I turn ambiguous product ideas into systems people can run, inspect, test, and maintain: APIs, persistence, integrations, evaluation loops, automation, and failure handling.

Featured work

Project What it demonstrates
Food Journal AI Bot Self-hosted Python/FastAPI/PostgreSQL service with LLM tool use, multi-channel input, idempotent webhooks, reliable delivery, and security-focused boundaries.
HeapHammer Java/JVM performance engineering through deterministic Minecraft server workloads, retained-memory diagnostics, and reproducible regression reports.
LeetCode Coach Service Python/FastAPI/SQLModel service with scheduled workflows, LLM coaching, migrations, a Telegram adapter, and a substantial automated test suite.
Skillweft Node.js CLI for progressive disclosure, trusted capability selection, immutable skill snapshots, and explicit agent-host mutations.
Identify the Author NLP experimentation with TF-IDF, classical models, calibration, cross-validation, and ensemble/submission tooling.
Value Sniper Experimental Python quantitative-research dashboard for multi-signal analysis and historical backtesting, with limitations documented rather than hidden.

More work

Trinity Pilot explores typed desktop/browser automation with evidence checks and bounded retries. Scriptorium explores schema-gated editorial workflows and agent coordination. Squadron packages structured multi-agent delegation and state tracking through MCP.

What I’m open to

Backend and AI engineering roles, product engineering, and focused automation work. If you need help turning an AI or workflow idea into a tested, maintainable system, connect with me on LinkedIn or email me.

Pinned Loading

  1. calorie-intake-tracker-ai-bot calorie-intake-tracker-ai-bot Public

    Production-oriented Python/FastAPI AI food journal with PostgreSQL, LLM agents, reliable webhooks, security controls, and automated testing.

    Python 1

  2. heaphammer heaphammer Public

    Deterministic Minecraft/JVM stress testing and retained-memory regression diagnostics for server-side workloads.

    Java 1

  3. leetcode-coach-service leetcode-coach-service Public

    Python/FastAPI LeetCode coaching service with PostgreSQL, scheduled workflows, LLM agents, Testcontainers, and production deployment tooling.

    Python 1

  4. skillweft skillweft Public

    Security-conscious capability manager for AI coding agents using progressive disclosure and immutable skill snapshots.

    JavaScript 1

  5. Identify-the-author Identify-the-author Public

    NLP author-classification pipeline exploring classical features, transformers, cross-validation, calibration, and model ensembling.

    Python 1

  6. value-sniper value-sniper Public

    Experimental Python quantitative-research framework for multi-signal market analysis, historical backtesting, and ML-assisted forecasting.

    Python 2