Skip to content
Hemz.
0%

Work · 09 entries

Selected work.

Things I've shipped, chasing one question: how do you make AI hold up outside the demo? Each card links to code or a write-up on how it was built.

t-SNE · 9 projects · seed 2018

9 projects shown

  • Agents & Tools

    notionwiki · 2026

    NotionWiki

    A Notion wiki your AI assistant can understand

    An open-source CLI that turns a Notion workspace into a flat, indexed markdown wiki an AI assistant can actually reason over: immutable raw sources, agent-curated synthesis on top, OS-native scheduled sync, and a force-directed graph view.

    • AI Assistants
    • Notion API
    • Python
    • CLI
    → code(opens NotionWiki in a new tab)→ write-up(opens NotionWiki in a new tab)
  • Observability

    observent · 2026

    Observent

    Claude Code plugin for AI observability

    A Claude Code plugin and Agent Skills that wire observability into multi-agent AI apps in one command: it detects your agent framework, instruments traces to Phoenix, Langfuse, or SigNoz, and validates the span hierarchy end to end.

    • AI Observability
    • Claude Code
    • OpenTelemetry
    • Agents
    → code(opens Observent in a new tab)→ write-up(opens Observent in a new tab)
  • Agents & Tools

    harness-mcp · 2025

    Harness MCP

    Reusable private AI toolkit

    A private, reusable AI toolkit built on the Model Context Protocol: write a capability once and plug it into any MCP-aware client, without your data ever leaving your walls.

    • MCP
    • AI Tools
    • Python
    → write-up(opens Harness MCP in a new tab)
  • Agents & Tools

    bedrock-agentcore · 2025

    Amazon Bedrock AgentCore

    Taking AI agents to production on AWS

    A hands-on tour of Amazon Bedrock AgentCore, the managed runtime for deploying AI agents with memory, identity, and tool access, and what it takes to move an agent from laptop to production.

    • Amazon Bedrock
    • AgentCore
    • AI Agents
    → write-up(opens Amazon Bedrock AgentCore in a new tab)
  • MLOps & LLMOps

    llmops-platform · 2024

    LLMOps Platform

    Full LLM lifecycle on Databricks

    A full lifecycle platform for LLMs on Databricks: model versioning, evaluation gates, deployment automation, and monitoring. The paved road from experiment to production.

    • LLMOps
    • GenAI
    • Databricks
    • Python
    → write-up(opens LLMOps Platform in a new tab)
  • RAG & LLMs

    local-rag · 2024

    Local RAG Application

    Secure RAG with LangChain

    A production-grade RAG chatbot that never phones home. LangChain, Llama 3, Qdrant, and Redis with full chat history, built for sensitive data that must stay on local infrastructure.

    • RAG
    • LangChain
    • Llama 3
    • Qdrant
    • Redis
    → write-up(opens Local RAG Application in a new tab)
  • Vision & Audio

    foodsight · 2024

    FoodSight

    AI food recognition

    A food-recognition web app fine-tuned on EfficientNetV2 with the Food101 dataset. Point it at a plate and it names the dish in real time.

    • Deep Learning
    • Computer Vision
    • Streamlit
    → live(opens FoodSight in a new tab)
  • MLOps & LLMOps

    mlops-platform · 2023

    MLOps Platform

    Plug-and-play MLOps framework

    A plug-and-play MLOps framework with CI/CD on Azure DevOps: continuous retraining, automated deployment, and drift monitoring, built for enterprise-scale ML operations.

    • MLOps
    • Databricks
    • CI/CD
    • Azure DevOps
    → write-up(opens MLOps Platform in a new tab)
  • Vision & Audio

    ask-audio · 2023

    Ask Audio

    AI audio analysis

    Feed it hours of recordings and get back the parts that matter: transcription, key insights, and intelligent summaries distilled from raw audio.

    • NLP
    • Speech-to-text
    • Audio
    • Python
    → write-up(opens Ask Audio in a new tab)