Skip to content
danraidPublic

About

No description, website, or topics provided.

Resources

Stars

0 stars

Watchers

1 watching

Forks

Latest commit

Β 

History

14 Commits

Folders and files

NameName
Last commit message
Last commit date
Β 
Β 
Β 
Β 

Repository files navigation

Daniel Raaid

Senior GenAI Engineer Β· AI Tech Lead Β· Data Scientist Β· Author

Research-led AI engineering | Reliable systems | Applied innovation

Agentic AI LLM Engineering RAG and Context Engineering MLOps and Data

πŸ“ Lisbon, Portugal Β |Β  πŸ‡΅πŸ‡Ή πŸ‡§πŸ‡· Portuguese & Brazilian citizenship Β |Β  🌍 Remote collaborations

LinkedIn Portfolio Books on Amazon


Professional Profile

I am a Senior Generative AI Engineer, AI Tech Lead and technical author with over seven years of experience in artificial intelligence, data science and data engineering. I design and implement scalable AI solutions, covering the lifecycle from data integration and model development through to cloud deployment, operational monitoring and continuous improvement.

My work focuses on agentic AI, large language models (LLMs), retrieval-augmented generation (RAG), context engineering and MLOps. I apply software engineering principles to the development of reliable, measurable and maintainable AI systems, with particular attention to evaluation, observability, security and operational efficiency.

I hold a master's degree in Bioinformatics and postgraduate qualifications in Software Engineering and Data Engineering. My previous experience in higher education complements my engineering practice with a research-oriented and methodical approach to complex technical challenges.

As an author, I publish technical books on context compaction, cache engineering and evaluation-driven development, supported by openly available practical exercises and reference materials.

Areas of expertise

  • Agentic AI and orchestration: multi-agent architectures, tool integration, workflow design and operational controls.
  • RAG and knowledge engineering: ingestion, embeddings, vector search, retrieval quality and evaluation.
  • Context and cache engineering: agent memory, context compaction, inference optimisation and information integrity.
  • AI evaluation and reliability: benchmarks, regression testing, observability, reproducibility and continuous quality assessment.
  • Cloud-native ML and data systems: Azure, AWS, Google Cloud, Python services, MLOps and data pipelines.
  • Technical research and education: evidence-based engineering methods, technical publications and open-source learning resources.

Technical Stack

Programming & Data

AI & Machine Learning

Vector Databases

BI & Data Visualization

Cloud & MLOps

Agentic AI, Orchestration & Retrieval

LangGraph Semantic Kernel RAG Azure AI Search PGVector PostgreSQL Neo4j

Data Platforms & Engineering

Apache Spark Apache Kafka PostgreSQL MongoDB SQL ETL / ELT

DevOps, Observability & Delivery

Git GitHub GitHub Actions Kubernetes OpenTelemetry CI/CD MLOps LLMOps

AI Engineering Methodology

My engineering approach follows a structured lifecycle: define measurable objectives and system requirements, establish the architecture, implement data and AI components, validate performance against explicit evaluation criteria, and deploy with appropriate monitoring and feedback mechanisms.

flowchart LR
    A["πŸ”Ž Research & discovery"] --> B["πŸ“ Design & specification"]
    B --> C["πŸ› οΈ Data, models & agents"]
    C --> D["πŸ§ͺ Evals & validation"]
    D --> E["πŸš€ CI/CD & deployment"]
    E --> F["πŸ“Š Monitoring & improvement"]
    F -->|"Feedback loop"| B
Loading
Engineering practice Primary objectives
πŸ”Ž Research-led design Defining requirements, constraints and evidence before building
🧩 Agent and RAG engineering Retrieval quality, tool integration, memory, reliability and traceability
πŸ§ͺ Eval-driven delivery Quality gates, regression tests, measurable outcomes and failure analysis
🐳 Production operations Containers, cloud services, CI/CD, logging, monitoring and iteration
πŸ” Responsible AI Privacy, human oversight, access control and auditable decisions

Published Books

My technical publications are available on Amazon in paperback and Kindle formats. Companion repositories provide open-access exercises, code and laboratory materials.

πŸ“˜ Context Compaction in Agent Harness Engineering

Theory, Mechanisms and Production Engineering

An engineering guide to context lifecycles, compression, memory, information preservation, evaluation and the operation of long-running AI agents.

Paperback Β· Kindle Β· Free exercises and code

πŸ“— Cache Engineering for AI Agent Harnesses

Design, Correctness and Operations for Reuse in LLM-Powered Agents

A practical and theoretical treatment of prompt, KV, semantic, retrieval and tool caching, with an emphasis on correctness, reuse and production behaviour.

Paperback Β· Kindle Β· Free exercises and code

πŸ“™ Eval-Driven Development

Engineering Reliable AI Systems Through Evidence

A research-informed approach to continuous AI evaluation, measurable quality, testing and evidence-based software development.

Paperback Β· Kindle Β· Free exercises and code

πŸ“• Agentic AI Engineering β€” Volume 01

Volume 01 of the Agentic AI Engineering series: architecture, implementation and evaluation of agentic AI systems.

Volume 01 Paperback Β· Free laboratories, Vol. 1 Β· Vol. 2 Β· Vol. 3 Β· Vol. 4

πŸ“’ Spec-Driven Development

Specification-led engineering, from explicit requirements to implementation and validation.

Paperback Β· Free exercises and code

Fiction published under the pen name Dan Raaid: Os Dois NΓ³s (Kindle).

Publication links above refer to Amazon listings verified in November 2026. Companion repositories provide independent learning materials; no purchase is required to use them.


Research & Experimental AI Engineering

I investigate the reliability of agentic AI through reproducible experiments, controlled evaluations and research-oriented prototypes. My current themes include memory integrity, multi-agent failure propagation and evidence-grounded retrieval. The research repositories document their protocols, limitations and validation stages; exploratory observations are not presented as confirmed findings.

Topics: Agent Memory Β· Context Compaction Β· Constraint Preservation Β· AI Safety

Memory integrity research

Abstract: Investigates whether repeated LLM-driven context compaction preserves evolving authorised constraints when agents make subsequent decisions. Compares full-history, recent-window, recursive-summary and pinned-constraint strategies using controlled tasks. The repository distinguishes protocol and instrument validation from confirmatory scientific evidence.

Topics: Multi-Agent Systems Β· Error Propagation Β· Verification Β· Fault Injection

Multi-agent reliability research

Abstract: Studies how incorrect observations and contaminated tool messages can propagate through agent interactions and affect final task outcomes. Explores independent verification, message quarantine and checker-feedback mechanisms across controlled exploratory tasks. Findings remain bounded by the tested environments and research stage.

Topics: Retrieval-Augmented Generation Β· Embeddings Β· Vector Search Β· Grounded Answers

Retrieval engineering research

Abstract: An applied engineering prototype for source-grounded question answering over documents. Implements ingestion, chunking, embedding-based FAISS retrieval and a FastAPI query interface, providing a foundation for evaluating retrieval relevance, traceability and future LLM integration.


Selected Open-Source Projects

Project Focus
RAG Knowledge Assistant Retrieval-augmented knowledge assistant and document-aware AI
Azure AI Agents MVP AI agents and orchestration on Azure
Azure Intelligent Document Document intelligence, data extraction and AI pipelines
Movie recommender with Neo4j and MLOps Graph data and applied machine learning
Data Engineering Portfolio Data platforms, pipelines and engineering examples
DNA Analysis Computational and bioinformatics-oriented work

Professional Experience

Period Role Organisation
2026–present Senior GenAI and Machine Learning Engineer BySix / Mozantech
Since 2025 Senior AI Engineer and AI Tech Lead D-Plan One
2019–2025 Senior Data Scientist and Consultant INFIN
2023–2025 Partner and Investment Advisor Saron Investments
2007–2021 University lecturer, academic coordination and leadership roles UniNorte and Faculdade EstΓ‘cio do Amazonas

My professional experience encompasses agent-based systems, AI-enabled applications, financial and operational analytics, natural language processing, predictive modelling, business intelligence, cloud architecture and end-to-end technical delivery.


Academic Background

  • Postgraduate Specialisation, Software Engineering, UNOPAR (2025)
  • Postgraduate Specialisation, Data Engineering, UNOPAR (2025)
  • Bachelor's, Data Science, UniFavip Wyden (2025)
  • Bachelor's, Financial Management, UniFavip Wyden (2024)
  • Master's, Bioinformatics (Genetics), INPA (2006)
  • Bachelor's, Bioinformatics, UFAM (2004)

Professional Certifications


GitHub Activity


Contact and Collaboration

I welcome professional collaboration on agentic AI, production LLM systems, applied research, AI evaluation and technical education. Please use the channels below to discuss relevant projects or opportunities.


Research-led design Β· Evidence-driven evaluation Β· Production-ready AI systems

About

No description, website, or topics provided.

Resources

Stars

0 stars

Watchers

1 watching

Forks

Releases

Packages

Contributors