Lolézio Viora Marquet

1. Software Development Engineer at Amazon

   September 2026 – Present · Prime Video · London, UK

   On the Prime Video Detail Page team, shipping customer-facing features to production for millions of customers, with AI coding agents as my default workflow. Joined full-time on a return offer from my Amazon Fuse internship.

   • Shipping customer-facing features to production on the Prime Video Detail Page, the page millions of customers see before they press play.
   • Built custom agentic coding harnesses, tailored to my team's development process, that write and review code changes end to end, so my manual work is steering and final sign-off.
   • The agents run as long-lived sessions on a remote cloud desktop with persistent access and automated workflows, so work keeps moving without my laptop attached.
   • TypeScript
   • Scala
   • React
   • AWS
   • AI coding agents
   • Automated code review

2. Software Development Engineer Intern at Amazon

   April 2025 – September 2025 · Amazon Fuse · London, UK

   Built an internal full-stack web application end to end, from user requirements to production, on a native AWS stack.

   • Shipped a dynamic React frontend against gathered user requirements, owning it from spec to launch.
   • Built an authentication system with gated APIs, so authorised users could query, edit and delete production database entries safely.
   • Backend natively on AWS: API Gateway, Lambda and DynamoDB, with no servers to operate.
   • 150 daily users and 3,000+ daily requests, peaking above 12,000, served at sub-300 ms latency and minimal operational cost.
   • Still in daily use a year after handover, with over 100,000 page views a quarter.
   • React
   • TypeScript
   • AWS Lambda
   • API Gateway
   • DynamoDB
   • CI/CD

3. Software Engineering & Data Science Intern at Equinor

   June 2024 – August 2024 · Renewables Bid Excellence · London, UK

   Built a cloud-hosted RAG system that turned hours of manual document search into a few-second query, for a team bidding on offshore wind farms.

   • NLP retrieval to surface the relevant documents for a question out of thousands of candidates.
   • An LLM on top of retrieval generating grounded answers that cite their sources, so bid teams could verify every claim.
   • A custom Selenium scraper that dynamically loads pages and extracts content to build the document database.
   • Deployed with Docker and Kubernetes, behind a chatbot-style web interface.
   • Python
   • NLP
   • RAG
   • LLMs
   • Selenium
   • Docker
   • Kubernetes

4. Software & Machine Learning Engineer at Maisha Design

   November 2023 – January 2026 · London, UK

   Two-year engagement applying machine learning to the creative workflow of an interior design firm, and building the software around it.

   • Generative AI and conditional GANs (CGANs) applied to the creative design process.
   • A full-stack application centralising supplier furniture catalogues in one database.
   • A project-management frontend generating invoices and letting designers browse supplier offerings.
   • Automated scrapers with LLM-enhanced category matching, keeping the catalogue current.
   • Cut project turnaround time by 30%.
   • Python
   • PyTorch
   • CGANs
   • LLMs
   • React
   • SQL
   • Web scraping

• Imperial College London

  First Class Honours October 2022 – June 2026

  MEng Electronic and Information Engineering

  • Core engineering across mathematics, AI/ML, hardware, software, networks and databases.
  • Master's thesis on neuro-symbolic learning under multi-modal data, reaching a new state of the art. Paper under review at a leading AI conference.

• Vienna International School

  43 / 45 August 2015 – May 2022

  International Baccalaureate Diploma

  • Higher Level: Mathematics Analysis & Approaches (7/7), Physics (7/7), Chemistry (7/7).
  • Standard Level: Economics (7/7), Spanish (7/7), English Language & Literature (6/7).
  • Extended Essay in Physics. SAT 1560 (2021).

• Machine learning

  Research through to production

  • PyTorch
  • Computer vision
  • Neuro-symbolic AI
  • Transformers
  • CGANs
  • NLP & RAG
  • NumPy
  • pandas
  • Model deployment

• Low-level & performance

  Where the cycles actually go

  • C / C++ (STL, memory model)
  • Multithreading & RTOS
  • Triton
  • CUDA
  • GPU kernel optimisation
  • Compilers
  • Profiling
  • Embedded systems
  • RISC-V

• Hardware

  Digital design and silicon-adjacent work

  • SystemVerilog / Verilog
  • FPGA (PYNQ-Z1, DE10-Lite)
  • CPU architecture
  • Pipelining & caching
  • Arduino
  • Signal processing

• Product & cloud

  Shipping things people use

  • TypeScript
  • Scala
  • React & React Native
  • AWS (Lambda, API Gateway, DynamoDB, EC2)
  • Google Cloud
  • Docker & Kubernetes
  • SQL
  • Flask
  • CI/CD

• LLMs & agents

  The default way I work

  • AI coding agents
  • Custom agent harnesses
  • Automated code review
  • RAG pipelines
  • LLM-enhanced data matching
  • Long-running remote agent sessions

• Also

  Picked up along the way

  • Python
  • Java & JavaFX
  • Swift
  • MATLAB
  • Julia
  • Isabelle
  • Shell scripting
  • Git

Every project has a full write-up. This is the index, 17 in total.

1. October 2025 – June 2026 Neuro-Symbolic Learning under Multi-Modal Data Master's thesis: teaching models to be right for the right reasons →
2. January 2026 – March 2026 AutoFuser Automatic Triton kernel fusion for deep learning models →
3. September 2026 – Present Agentic Coding Harnesses at Prime Video Agents that write and review production code for a page millions of customers use →
4. July 2024 – Present Cryptocurrency Trading Bot A transformer trading live capital, 24/7, since 2024 →
5. April 2025 – September 2025 Amazon Fuse Internal Platform A serverless internal tool, from requirements to production →
6. February 2025 – March 2025 STM32 Synthesiser, running DOOM Real-time embedded C++, a full octave, and a 3D game engine in 64 kB →
7. June 2024 – August 2024 Offshore Wind RAG System Retrieval-augmented generation over thousands of bid documents →
8. May 2024 – June 2024 Game of Life on FPGA 200,000 evolutions per second, controlled by hand gestures →
9. January 2024 – March 2024 C90-to-RISC-V Compiler A working C compiler, written in C++ →
10. October 2023 – December 2023 Pipelined RISC-V CPU A 32-bit CPU in SystemVerilog, from single-cycle to pipelined →
11. January 2025 – February 2025 PiTrainer An AI personal trainer: Raspberry Pi, cloud, and a mobile app →
12. January 2023 – June 2024 Project ATLAS Leading 15 engineers to build an autonomous drone →
13. February 2024 – March 2024 2-Player Flight Simulator FPGA joysticks, an AWS server, and two planes in one world →
14. November 2023 – January 2026 Maisha Design Platform Generative AI and a full-stack platform for an interior design firm →
15. June 2023 Fyrryx Rover A wirelessly-driven rover on two microcontrollers →
16. 2021 Search-and-Rescue Quadcopter YOLOv3 obstacle avoidance for alpine rescue →
17. 2023 Autonomous Greenhouse Closed-loop environmental control to increase plant yield →