Leonel Olímpio Silima

Welcome

I'm Leonel Olímpio Silima

MLOps / LLMOps Engineer at LynxMind Portugal · PhD Researcher at FEUP · Junior Researcher at INESC TEC · Building production-grade AI systems for research and industry

Go to Projects Download CV
Portrait of Leonel Olímpio Silima

About

Hello, I’m Leonel Olímpio Silima

MLOps / LLMOps Engineer · PhD Researcher · AI Systems Engineer

I work as an MLOps and LLMOps Engineer at LynxMind Portugal, where I have been building and operating production-grade AI systems for more than three years. My work focuses on cloud-native platforms, scalable ML/LLM architectures, deployment automation, monitoring, and continuous evaluation for real operating environments.

I hold a BSc in Computer Science from Universidade Católica de Moçambique and an MSc in Applied Informatics from Universidade de Aveiro, developed in collaboration with Aily Labs in Barcelona. I am currently a PhD Researcher in Electrical and Computer Engineering at FEUP and a Junior Researcher at INESC TEC within LIAAD.

My trajectory combines academic research and applied engineering across Europe and Asia, including China, with experience in multicultural and distributed teams. I am particularly interested in reproducible AI systems, MLOps observability, LLM evaluation, and reliable delivery of intelligent platforms.

My training also includes Deep Learning from Stanford University, Generative AI specialization at FCUP, and IBM Ethical Hacker certification, complementing my engineering background with strong foundations in modern AI methods and secure system design.

Download my CV

Services

MLOps and LLMOps pipeline architecture

MLOps / LLMOps Engineering

  • Airflow
  • MLflow
  • Kubeflow

Design and operation of ML and LLM systems across experimentation, orchestration, evaluation, serving, and lifecycle governance for production environments.

Cloud-native platform and infrastructure architecture

Cloud & Platform Engineering

  • AWS / Azure
  • Docker
  • Terraform

Scalable cloud-native platforms for AI and engineering workloads, with infrastructure as code, container orchestration, and resilient deployment patterns.

DevOps CI CD delivery and monitoring workflow

DevOps & CI/CD

  • GitHub Actions
  • GitOps
  • Release Pipelines

Continuous integration and delivery pipelines for applications and AI platforms, with automated validation, promotion strategies, and release reliability.

Applied AI solution and data workflow

AI & Data Solutions

  • scikit-learn
  • Deep Learning
  • Computer Vision

Applied AI systems for research and industry, combining data pipelines, feature engineering, model development, experimentation, and operational delivery.

Monitoring, telemetry and secure delivery dashboard

Observability & Secure Delivery

  • Prometheus
  • Grafana
  • OpenTelemetry

Monitoring, telemetry, traceability, and security-aware operations for AI and cloud workloads, improving reliability, visibility, and controlled delivery.

Systems operations and engineering support

Systems & Operations

  • Linux
  • Networking
  • Automation

Production operations, troubleshooting, systems administration, and environment support for reliable services and engineering teams.

Technology Stack

MLOps / LLMOps

  • Kubeflow
  • MLflow
  • Airflow
  • TFX
  • LangChain
  • LlamaIndex
  • BentoML
  • Feast
  • Weights & Biases
  • Prompt Engineering
  • RAG
  • Model Evaluation

Cloud & Platform

  • AWS
  • Azure
  • Docker
  • Kubernetes
  • Terraform
  • GitHub Actions
  • Linux
  • Platform Automation
  • Containerization

Languages, APIs & Prototyping

  • Python
  • Go
  • Bash
  • FastAPI
  • REST APIs
  • SQL
  • Streamlit
  • Jupyter
  • Gradio
  • scikit-learn

Observability

  • Prometheus
  • Grafana
  • OpenTelemetry
  • Logging
  • Tracing
  • Monitoring Dashboards
  • Alerting

Systems Engineering

  • Distributed Systems
  • Computer Vision
  • DevOps
  • CI/CD
  • Networking
  • Secure Delivery

Research Focus

  • MLOps Observability
  • LLM Evaluation
  • Reproducible AI
  • AI Systems Engineering
  • Generative AI
  • Industry-grade AI Platforms

Certifications

Design Thinking for Human-Centered IoT Solutions

LearnQuest · Coursera · 2026

Verification link

Deep Learning Specialization

Stanford University / deeplearning.ai

Verification link

Generative AI Specialization

FCUP · University of Porto

Verification link

IBM Ethical Hacker Certificate

IBM

Verification link

AWS Machine Learning

Amazon Web Services

Verification link

Microsoft Azure Cloud

Microsoft

Verification link

Research Interests

Research Interests

MLOps Observability, LLM Evaluation, Reproducible AI, AI Systems Engineering, production reliability for intelligent systems, and cloud-native platforms for research and industry.

Public Research Profiles

ORCID: 0000-0003-0160-5979

Speaker profile: sessionize.com/leonel-silima

Impact Snapshot

3+ years delivering production AI systems in industry, combining research, deployment automation, observability, and engineering workflows across academic and operational environments.

Selected Projects

Production RAG and retrieval architecture

GenAI Knowledge Platform with RAG

Built a production-grade retrieval and generation platform for enterprise knowledge workflows, with ingestion, evaluation, response tracing, and deployment-ready serving patterns.

Technologies: LangChain, LlamaIndex, FastAPI, Vector Search, Docker, Kubernetes

Impact: Structured GenAI delivery around reproducibility, observability, and controlled rollout.

GenAI agents and orchestration workflow

Multi-Agent GenAI Orchestration

Designed a robust agent-based workflow for task decomposition, tool use, prompt routing, and response validation in complex AI-assisted processes.

Technologies: LangChain, Prompt Engineering, FastAPI, Python, Evaluation Pipelines

Impact: Advanced prototyping and operationalization of complex Generative AI workflows.

MLOps platform on Kubernetes with pipelines and registry

MLOps Platform on Kubernetes

Built an end-to-end MLOps platform for training, tracking, versioning, serving, and release control across model lifecycle stages.

Technologies: Kubeflow, MLflow, Airflow, Docker, Kubernetes, Terraform

Impact: Reduced manual model promotion steps and improved consistency across delivery environments.

ML observability and telemetry dashboards

MLOps Observability & Evaluation Stack

Implemented monitoring, tracing, and evaluation layers for model and platform behavior, with visibility into performance, drift signals, and operational health.

Technologies: Prometheus, Grafana, OpenTelemetry, MLflow, Python

Impact: Improved troubleshooting speed and observability maturity in AI operations.

GitOps based delivery workflow

GitOps Delivery Workflow

Designed a GitOps-oriented deployment model with environment promotion, automated validation, rollback readiness, and version-controlled infrastructure changes.

Technologies: GitHub Actions, Kubernetes, Terraform, Bash, Docker

Impact: Strengthened release governance and deployment traceability.

Cloud DevOps platform architecture

Cloud-Native DevOps Platform

Engineered a cloud-native platform for application and AI workloads with reusable CI/CD components, infrastructure automation, and environment standardization.

Technologies: AWS, Azure, Docker, Kubernetes, GitHub Actions, Terraform

Impact: Increased delivery consistency and reduced configuration drift.

Secure DevOps and observability architecture

Secure Delivery & Runtime Hardening

Combined secure delivery practices with runtime visibility, hardening, and operational controls for services deployed in cloud-native environments.

Technologies: Linux, CI/CD, Monitoring, Security Controls, Observability Tooling

Impact: Improved deployment confidence and operational resilience.

Contact

Email: leonel.silima0@gmail.com