Skip to content
View VDurocher's full-sized avatar
1 follower · 2 following

Achievements

Achievement: Pull SharkAchievement: Pair ExtraordinaireAchievement: Quickdraw

Achievements

Achievement: Pull SharkAchievement: Pair ExtraordinaireAchievement: Quickdraw

Block or report VDurocher

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.

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
VDurocher/README.md

Victor Durocher

Full-stack developer building AI tooling, iOS packages, and indie SaaS.
I focus on production-grade systems: LLM evaluation, RAG pipelines, and native Swift AI integration.

Stack

TypeScript Python Swift Next.js Supabase OpenAI Anthropic

Projects

Project Description Stack
Swift AI Agent Core Swift 6 package for integrating LLM agents into iOS/macOS apps — streaming, retry, SwiftData persistence, zero dependencies Swift
LLM Benchmarker Suite Production evaluation toolkit — 6 evaluators, LLM-as-judge, multi-model comparison, CI/CD ready Python
RAG Knowledge Assistant Turn documents into a cited knowledge base — FAISS, hybrid search, fully local or OpenAI Python
Prompt Evolution Engine Genetic algorithm that automatically evolves prompts — generates variants, scores them, iterates Python
AI Email Intent Analyzer Production pipeline — classifies support emails, extracts entities, drafts replies in one API call Python

Pinned Loading

  1. Swift-AI-Agent-Core Swift-AI-Agent-Core Public

    Professional Swift package for integrating LLM agents into iOS apps with production-grade reliability. Features streaming support, retry mechanisms, and token management.

    Swift 3

  2. RAG-Knowledge-Assistant RAG-Knowledge-Assistant Public

    Production-ready RAG system for SMEs — turn your documents into a cited, hallucination-free knowledge base. Supports OpenAI or fully local Ollama+HuggingFace.

    Python

  3. LLM-Benchmarker-Suite LLM-Benchmarker-Suite Public

    Professional evaluation toolkit for LLM outputs — logic, format, and hallucination checks for production-grade reliability

    Python

  4. prompt-evolution-engine prompt-evolution-engine Public

    Automatically evolves prompts using a genetic algorithm — generates variants, scores them with LLM-as-judge, and iterates until the optimal prompt emerges.

    Python