Skip to content

Instantly share code, notes, and snippets.

@Siddhant-K-code

Siddhant-K-code/my-random-ai-projects.md

Last active October 5, 2025 17:27
Show Gist options
  • Save Siddhant-K-code/5de9207adeeba88299e81408422c9dfb to your computer and use it in GitHub Desktop.
Save Siddhant-K-code/5de9207adeeba88299e81408422c9dfb to your computer and use it in GitHub Desktop.
Download ZIP
Random AI projects of Siddhant
Raw
my-random-ai-projects.md

👋 About Me

Hi, I'm Siddhant Khare, a software engineer building developer tools, distributed systems, and fun side projects with AI.

🚀 AI Projects

A collection of interesting / fun / random experiments (not in priority order).

  • A2AS Implementation
    Proof-of-concept implementation of the A2AS (Agent-to-Agent Security) framework demonstrating Behavior Certificates, Authenticated Prompts, Security Boundaries, and Enforcement Gates. Inspired by the A2AS Paper.

  • Agentflow
    Kubernetes for AI agents | One single platform combining orchestration runtime, prompt operations, security layer, observability stack, and cost-aware scheduling. Deploy production AI workflows with confidence: DAG execution, multi-provider routing, budget enforcement, and semantic tracing built-in.

  • KV-Cache Profiler
    Profiler for LLM inference that analyzes and visualizes KV-cache memory usage across transformer layers. Helps identify inefficiencies, optimize memory, and understand sequence length scaling.

  • TokenVM
    TokenVM is a high-performance runtime that treats LLM KV cache and activations as a virtual memory working set across GPU VRAM → pinned host RAM → NVMe storage, with intelligent paging, prefetching, and compute-copy overlap.

  • CloudArb
    GPU arbitrage platform that optimizes cloud compute costs for AI companies through real-time multi-cloud allocation and automated deployment.

  • LLMTraceFX
    GPU-level LLM execution profiler on Modal. Goes beyond flame graphs with kernel breakdowns, stall attribution, cache diagnostics, and GPT-powered insights.

  • Song Vector Explorer
    Explore song lyrics as interactive 3D vector spaces using OpenAI embeddings + TensorBoard visualization.

  • actionsec
    Fast, local-first CLI to scan GitHub Actions workflows for security vulnerabilities using AI.

  • SageMap · Live Demo
    Interactive tool to map and evolve personal beliefs using OpenAI.

  • ArchiFusion · Live Demo
    Transforms architectural ideas into 3D building models with AI.

  • LLM Parallelism Explorer
    Proof-of-concept for exploring parallelism in LLM inference. Accompanies my blog post on execution parallelism.

  • AI Agent Orchestrator
    Multi-agent AI system on Cloudflare Workers + Containers that enables collaboration across specialized AI agents for research and analysis.

  • MediBrief
    Automates summarization of PubMed research papers into concise narrated videos, auto-published on YouTube.

  • Cloud Architect AI
    AI-assisted design tool for visualizing, planning, and implementing cloud infrastructure.

  • LangChain × OpenAI: Bring Your Own Data

  • Contextify
    Lightweight Python script to streamline injecting project context into LLM prompts.

🔌 MCP Servers

  • Apple Notes MCP Server
    Interact with Apple Notes via natural language — create, search, and retrieve notes seamlessly with Claude or other MCP-compatible assistants.

  • Smart Photo Journal MCP Server
    Search and analyze your photo library with AI. Useful for reminiscing or finding specific memories.

  • Devcontainer MCP Server
    Manage DevContainers using AI prompts in VS Code, Cursor, or any MCP-compatible editor.

📝 Selected Blogs

AI-related posts:

📚 Full archive: dev.to/siddhantkcode

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment