kgkazakos (KostasK) · GitHub
Skip to content
View kgkazakos's full-sized avatar
2 followers · 1 following

Block or report kgkazakos

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. 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
kgkazakos/README.md

Kostas Kazakos

Computational Product Researcher · PhD in HCI · Melbourne, AU

I architect & build AI-native research infrastructure — systems that keep humans in command while extending the reach of qualitative and quantitative insight.

Computational Product Research (CPR)

CPR is the bridge from static insight to agentic intelligence. I architect open-source tools using Python, LangChain, DSPy, and LLMs that make research rigorous, transparent, and scalable.

Project What it does Status
CausalTrack Multimodal Say-Do Gap detection · Gemini + Neo4j · 611-node knowledge graph Live
Synthetic User Council Agent-led ecosystem simulation for pre-deployment product research Live
Agentic Payment Trust Mapping Maps the Trust Line for agentic payment feature decisions Live
Cognitive Load Monitor Measures cognitive load from interaction telemetry and interprets the source using a CLT-grounded AI agent Live
Research Conductor AI agent that conducts user research studies autonomously, with static config-driven gates defining when human oversight is required. Live
Parallax Spatial research canvas that compares reasoning models against standard LLMs for critiquing qualitative clustering, making the AI's chain-of-thought visible to researchers so they can see why the AI disagrees — not just that it does. Live
Ethos Encoding research methodology as alignment principles for autonomous research agents. Live
UXR Arena Benchmarks AI tools on 10 standardised UX research tasks, scoring against Ethos's constitutional principles with dual evaluation — automated compliance and expert pairwise ranking. In Progress

40 orbital tools shipping almost daily through H1 2026 · full roadmap →

Stack

Python LangChain DSPy LangGraph Neo4j Streamlit Gemini Claude OpenAI pytest GitHub Actions Pandas NetworkX Click

Writing

Currently writing The AI-Native Product Researcher: Scaling Empathy and Rigor — a book on the transition from episodic insights to continuous intelligence.

Pinned Loading

  1. causaltrack causaltrack Public

    Audio-based behavioral truth detection for user research - A methodology validation study

  2. synthetic-user-council synthetic-user-council Public

    A multi-agent AI system that challenges researcher assumptions before user studies begin.

    TypeScript

  3. research-conductor-public research-conductor-public Public

    AI agent that conducts user research studies autonomously, with static config-driven gates defining when human oversight is required.

    Python

  4. parallax-public parallax-public Public

    Parallax: Co-Reasoning Research Canvas — Architecture & Research Design

  5. ethos-public ethos-public Public

    Ethos: Encoding research methodology as constitutional alignment principles for autonomous research agents. 5 meta-principles, 19 rules, structured deliberation. P6 in the Computational Product Res…

    Python

  6. uxr-arena-public uxr-arena-public Public

    A domain-specific benchmark for evaluating AI tools on UX research tasks — 10 standardised tasks, constitutional compliance scoring, and Bradley-Terry rankings.