CV

About Me

Driven and passionate, I find joy in every aspect of software development, from initial planning to the final polishing stage. Whether in pursuit of my computer science and engineering degree or through freelance projects, I have consistently taken on the responsibility of organizing and leading teams, guiding them towards successful project outcomes. While steering the team towards the goal, I have always remained hands-on, offering assistance to fellow developers and tackling tasks independently when needed. This approach allows me to gain a comprehensive understanding of the entire project, a critical aspect during the documentation stages. My unwavering interest in the field compels me to seek knowledge through collaborative efforts and personal research, enabling me to be a reliable source of help to others and a valuable asset to any team I join.

Work Experience

Equistamp

• Contractor Research Engineer
• March 2025 – Current
• Built and maintained evaluation and task scaffolding for LLM agents; authored tasks and settings used in production evaluations
• ControlArena (UK Government): 13 PRs across control-arena and ca-k8s-infra over 2 tasks (ideation, solution implementation, scorer implementation); research on task ideation (Mar 14 – Jun 11, 2025)
• LinuxBench (Redwood Research): 12 PRs spanning setting development, task ideation/solutions/scorers, and evaluation-suite tooling (Jun 4 – Aug 12, 2025; ongoing)
• Reworked a setting to add complexity; implemented 2 main and 2 side tasks; provided baseline solutions for 2 task-combination variants
• Helped METR migrate evaluation workflows from Vivaria to an inspect-based platform; contributed to testing and rollout

METR

• Model Evaluation and Task Engineer
• January 2024 – March 2025
• Ran human expert baselines and graded LLM agent runs across multiple task suites; developed and fixed tasks for reliability
• Authored evaluation protocols and pass/fail criteria for large language model (LLM) assessment; acknowledged in HCAST

Softinsa - IBM - Fundão, Portugal

• Senior Software Developer
• June 2021 – March 2025
• Started as an intern and progressed to full-time developer, eventually promoted to Senior Software Developer
• Developed and maintained multiple backend systems and web applications using Python, Flask, and FastAPI
• Built automated knowledge test generation platform for internal candidate evaluation, including admin dashboard and candidate testing interface
• Contributed to open-source utility functions and backend systems for large-scale IBM projects
• Worked with technologies including Python, FastAPI, Flask, MongoDB, MSSQL, Docker, Kubernetes, and OpenShift

Zirak - Mondovi, Italy

• Technical Lead, Software Developer, Software Tester

• September 2019 – September 2020

  Led the development of a prototype for a blockchain-based supply chain traceability system. This system certified product origins and utilized machine learning models, including time-series predictions, to optimize product flow within the supply chain.

Education and Training

Master Degree in Informatics Engineering

• 2018 – 2020
• Universidade da Beira Interior, Covilhã, Portugal
• Project: “Certitex: a Textile Certified Supply Chain” — Advisor: Prof. Luís A. Alexandre.
• Member of the NOVALINCS research group.
• Overall score: 16/20.

Bachelor Degree in Informatics Engineering

• 2015 – 2018
• Universidade da Beira Interior, Covilhã, Portugal
• Project: “Something Brawl” — Advisor: Prof. Frutuoso Gomes.
• Member of the ReGain group.
• Overall score: 13/20.

Technologies

Programming languages

Technologies and Tools

Language Skills

• Portuguese (Mother tongue)
• English (C2 Listening & Reading, C1 Spoken Production & Spoken Interaction, C1 Writing)
• Spanish (B1 Listening & Reading, A2 Spoken Production & Spoken Interaction, A2 Writing)