Skip to content
View DanishShaikh18's full-sized avatar
15 followers · 4 following

Achievements

Achievement: Pull SharkAchievement: YOLOAchievement: Quickdraw

Achievements

Achievement: Pull SharkAchievement: YOLOAchievement: Quickdraw

Block or report DanishShaikh18

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
DanishShaikh18/README.md

Danish Shaikh

I build AI systems that reason and act, not just respond autonomous agents, retrieval pipelines, fine-tuned models, and the cloud infrastructure that runs them. My work spans healthcare, cybersecurity, and developer tooling, with a focus on understanding why an architecture works, not just shipping a demo.

🚀 Featured Projects

☁️ Fine-Tuned Log-to-JSON Parser

Fine-tuned Llama-3.2-1B using QLoRA, quantized with GGUF + llama.cpp, and deployed on Google Cloud Run for CPU-only, serverless log parsing and PII redaction cost-engineered end to end, from training to a scale-to-zero production deployment.

🤖 Agentic QA Fuzzer

Autonomous ReAct agent built in LangGraph, using the Model Context Protocol (MCP) to isolate reasoning from tool execution. Parses OpenAPI schemas, mutates payloads to surface unhandled server exceptions, and auto-generates pytest regression tests for every discovered bug.

📚 Multi-Agent RAG System

A 5-agent RAG pipeline in LangGraph Guardrail, Research, Retrieval, Response, and Validation agents working together to deliver grounded, source-attributed answers over PDF documents using FAISS semantic search.

🧪 LLM Evaluation Pipeline

An LLM-as-a-Judge framework for benchmarking RAG system outputs on correctness, completeness, relevance, and hallucination risk.

🤖 AI & LLM

Agentic AI Multi-Agent Systems RAG Fine-Tuning LLM Evaluation AI Guardrails Tool Calling

⚙️ Frameworks & Tools

LangGraph LangChain MCP Google ADK Gemini API FAISS llama.cpp Unsloth

☁️ Cloud & Backend

Google Cloud Vertex AI Agent Platform Cloud Run FastAPI Docker PostgreSQL

💻 Languages

Python SQL Git

🌐 Connect

Pinned Loading

  1. Fine-Tuned-Log-to-JSON-Parser Fine-Tuned-Log-to-JSON-Parser Public

    Cut raw cloud logs down to clean, PII-safe structured data running entirely serverless, at near-zero cost, with zero GPU.

    Python

  2. MCP-Fuzzer-Agent MCP-Fuzzer-Agent Public

    Finds bugs humans miss by reasoning like a QA engineer then writes its own regression tests so they never come back.

    Python

  3. Multi-Agent-RAG-System Multi-Agent-RAG-System Public

    Answers grounded in your documents, not the model's guesses every claim traceable back to its source.

    Python

  4. LLM-Evaluation-Pipeline LLM-Evaluation-Pipeline Public

    Answers the question every RAG team avoids: is your system actually right, or does it just sound right?

    Python