Skip to content
View resul-ai's full-sized avatar
🦒
The Harnessing of Artificial Intelligence
🦒
The Harnessing of Artificial Intelligence

Block or report resul-ai

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. 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
resul-ai/README.md

Resul Caliskan

AI engineer · Agentic systems, language models, and applied NLP

I work on the software around language models: tools, context, memory, structured interfaces, evaluation, and safety boundaries. The interesting part is not merely calling a model, but shaping the harness that lets it reason and act reliably.

Giraffe Analytics is where I turn these ideas into focused, production systems.

Working on

  • agentic programming patterns and tool-using systems
  • model harnesses: context, memory, structured outputs, and evaluation loops
  • reliable execution boundaries and human-in-the-loop controls
  • applied language systems, including healthcare NLP

Featured open source

A dependency-free Python policy gate for AI agent tool calls. It produces auditable allow, deny, or confirm decisions before a tool executes, with checks for filesystem scope, network destinations, secrets, and destructive operations.

Healthcare NLP pipeline examples and custom named-entity recognition training.

How I build

model + context + tools + feedback
                ↓
          useful behavior

Good agentic software is largely interface design: what the model can see, what it can do, how its work is checked, and where people remain in control. I like small reproducible systems, typed boundaries, measurable behavior, and clear failure modes.

Connect

Pinned Loading

  1. qwen38-27b qwen38-27b Public

    Qwen3.8-27B FP8 on RunPod vLLM — OpenAI-compatible agent endpoint

    Python 1 1

  2. rag-with-gpt5 rag-with-gpt5 Public

    Production-ready RAG API with GPT-5 nano, FastAPI, PostgreSQL, and pgvector.

    Python 1

  3. agent-tool-safety agent-tool-safety Public

    Deterministic allow, deny, and human-confirmation policies for AI agent tool calls.

    Python 1

  4. rag-chatbot rag-chatbot Public

    Document-grounded RAG chatbot with Streamlit, Ollama, FAISS, Redis, and Docker.

    Python 1

  5. SparkNLP-CustomNER SparkNLP-CustomNER Public

    Healthcare NER pipeline combining Spark NLP models and training a custom clinical NER model.

    Jupyter Notebook 1

  6. QA-Chatbot QA-Chatbot Public

    Document question-answering chatbot built with OpenAI, Streamlit, and Python.

    Python 1 1