Skip to content
View danteprz9621's full-sized avatar
🥨
Working from home
🥨
Working from home
  • Mexico
  • 06:12 (UTC -06:00)

Block or report danteprz9621

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
danteprz9621/README.md

Hi, I'm Dante 👋

SDET focused on test automation and AI/LLM evaluation — building the tooling that catches bugs and bad model behavior before your users do.

🧰 Mobile/Web E2E · UI · API testing

🌎 Mexico · 🥨 Working from home

What I work on

I build evaluation, observability, and red-teaming pipelines for LLM-based agents — mostly around a fictional "Trailhead Travel" support bot I use as a testbed for these tools.

  • 🧪 trailhead-travel-agent-eval LLM-agent quality suite (single-turn Q&A, multi-turn chat, RAG) using DeepEval — correctness, safety, and hallucination metrics scored by a local judge model.

  • 📚 trailhead-travel-rag-eval RAG-pipeline evaluation using Ragas — faithfulness, context precision/recall, and answer relevancy, with noise-aware CI thresholds.

  • 📈 trailhead-travel-observability Tracing and CI regression gating for an LLM agent using LangSmith — versioned eval datasets, a GitHub Actions prompt-regression gate, and drift monitoring.

  • 🛡️ trailhead-travel-red-team Automated red-teaming and safety layering for a RAG agent — Promptfoo attack generation, Guardrails AI validation, Llama Guard classification, before/after break-rate comparison.

🔭 Currently building out UI/E2E test automation projects with Selenium and Playwright — stay tuned.

Toolbelt

Test Automation

Java Python Selenium Playwright GitHub Actions

AI/LLM Evaluation

DeepEval Ragas LangSmith Guardrails AI

Let's connect

LinkedIn

Pinned Loading

  1. trailhead-travel-agent-eval trailhead-travel-agent-eval Public

    LLM-agent quality suite for a fictional travel-support bot (single-turn Q&A, multi-turn chatbot, RAG) using DeepEval — correctness, safety, and hallucination metrics scored by a local judge model.

    Python

  2. trailhead-travel-rag-eval trailhead-travel-rag-eval Public

    RAG-pipeline evaluation for a fictional travel-support agent using Ragas — faithfulness, context precision/recall, and answer relevancy, with noise-aware CI thresholds.

    Python

  3. trailhead-travel-observability trailhead-travel-observability Public

    Observability and CI regression gating for an LLM agent using LangSmith — tracing, versioned eval datasets, a GitHub Actions prompt-regression gate, and production drift monitoring.

    Python

  4. trailhead-travel-red-team trailhead-travel-red-team Public

    Automated red-teaming and safety layering for a RAG agent — Promptfoo attack generation, Guardrails AI validation, and Llama Guard classification, with before/after break-rate comparison.

    Python