Skip to content
View AlexisMarasigan's full-sized avatar
8 followers · 15 following

Achievements

Achievement: Pull Sharkx4Achievement: Pair Extraordinairex4Achievement: QuickdrawAchievement: YOLO

Achievements

Achievement: Pull Sharkx4Achievement: Pair Extraordinairex4Achievement: QuickdrawAchievement: YOLO

Block or report AlexisMarasigan

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.

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
AlexisMarasigan/README.md

Hi, I'm Alexis 👋

AI Engineer · Stockholm

I build production LLM systems — agent runtimes, eval and observability, the infra in between. Mathematics BSc (La Sapienza, Rome) + MSc coursework in Reinforcement Learning and stochastic processes (Stockholm University).

Selected work

  • 🔧 Merged into Google's Agent Development Kit — adk-python PR #3756: diagnosed and fixed an httpx client-closure bug in VertexAiSessionService.get_session() during multi-page event pagination. Merged via Copybara, commit b725045.
  • ♟️ OOP-and-Functional-Chess — chess engine implemented twice (object-oriented and functional) to compare paradigms head-to-head.
  • 📊 bustub-vectordb — fork of CMU's BusTub DBMS, exploring vector-search internals.

Currently exploring

  • Eval-first agent development — closing the loop between agent decisions and business outcomes
  • Hybrid RAG (graph + document) and where it actually beats vector-only
  • Coding-agent harnesses (Claude Code, Cursor) for team velocity

Stack

AI / LLM · LangGraph · LangFuse · hybrid RAG (Neo4j + DynamoDB) · prompt + eval design Backend · Python (FastAPI) · TypeScript (Remix on Node) Infra · AWS · Docker · Kubernetes · CI/CD Math · Numerical analysis · probability · optimization · RL

Connect

LinkedIn · alexisjh.one · alexismarasigan31@gmail.com

🎻 Fun fact: principal second violinist in two Italian youth orchestras (2016–2022) before fully pivoting to engineering.

Alexis Marasigan — GitHub stats

Pinned Loading

  1. fastmcp-kit fastmcp-kit Public

    Framework for building authenticated, scoped, observable MCP servers on top of FastMCP. Tool registration, scoped auth keys, Prometheus/Grafana, multi-tenancy.

    Python 1

  2. OOP-and-Functional-Chess OOP-and-Functional-Chess Public

    Python

  3. google/adk-python google/adk-python Public

    An open-source, code-first Python toolkit for building, evaluating, and deploying sophisticated AI agents with flexibility and control.

    Python 20.1k 3.6k

  4. tadata-org/fastapi_mcp tadata-org/fastapi_mcp Public

    Expose your FastAPI endpoints as Model Context Protocol (MCP) tools, with Auth!

    Python 11.9k 952