Iñaki Arango

Documentation

About

I'm a senior at Harvard pursuing an AB/SM in CS and Stats. Before coming to college I took a gap year and a half to work as a software engineer and VC, and started a company that optimizes the recycling initiatives of governments and consumer brands.

Currently, I'm performing research under Dr. Marinka Zitnik at Harvard Medical School on the application of GNNs, LLMs, and ~agents~ to science.

Work Experience

H

Harvard Medical School

November 2024 - Present
AI Researcher @ Zitnik Lab
  • Led a team of 4 developing automated knowledge graph construction methods, producing a graph with 21M edges and 88M properties.
  • Collaborated with the Gates Foundation on methods for leveraging knowledge graphs in clinical settings, soon to undergo clinical trial.
R

Recycoin

January 2020 - Present
Co-founder
  • Led and closed negotiations with clients for contracts worth $650K+, including Procter & Gamble, L'Oréal and Unilever.
  • Designed architecture and implemented the 4 software components required for the MVP: 2 apps (React Native with shared company packages in monorepo), a management dashboard (Redwood, React, GraphQL), and server (Node, TypeScript, GraphQL, Prisma).
  • Hired and managed of 8 technical employees, including project managers, DevOps, and software engineers.
  • Bootstrapped the company into covering recycling planning services for 5,000,000 people, or 11% of the Argentinian population.
J

Jane Street

May 2025 - August 2025
Software Engineering Intern
  • Developed a new method for identifying pending refactors across the codebase.
  • Designed new distributed infrastructure for benchmarking trading strategies, placing top 10% in trading competition.
G

Google

May 2024 - August 2024
Software Engineering Intern
  • Implemented an on-device offline spell-checking ML model for Google Docs and Slides suporting 5 locales (Java, Closure, WASM).
  • Optimized dependency injection strategy for JavaScript web files, decreasing size of critical binaries by 3% and 4% (Closure).
A

Amazon

May 2023 - August 2023
Software Engineering Intern
  • Architected and implemented new service that automates the training and deployment of new machine learning models, reducing time to onboard secure data sources by 75%, time to deploy new models by 66%, and overall compute cost (TypeScript, CLI, AWS, SageMaker).
  • Presented technical and business impact of the service to 65+ engineers and managers within the Alexa Proactive organization.

Education

H

Harvard University

2022 - 2025
MA in Computer Science, BA in Computer Science with a Secondary in Statistics
  • Extracurriculars: Research (Zitnik Lab, Amin Lab), President of Harvard SHPE, Datamatch Algo Lead, Trading Competition Organizer (HUTC 2024), Harvard Quantitative Traders Club Director, Neo Scholar Finalist
  • Relevant Coursework: Probability (STAT 110, Teaching Fellow), Algorithmic Game Theory (CS 136, Research Assistant), Statistics (STAT 111), Advanced Algorithms (CS 124), Interpretable ML (CS 282), Systems Programming (CS 61), ML Networks (CS 243), Compositional AI Systems (CS 2821R), Advanced Topics in ML (CS 2822R), Advanced Topics in ML (CS 2823R)
  • Scholarships: John Harvard Scholarship 2023, John Harvard Scholarship 2025 (Awarded to top 5% of the class)
R

Rice University

2021 - 2022
BSCS Computer Science, Minor in Mathematics (Transferred)
  • Extracurriculars: Rice New Energy Fund (Head of Operations), Rice Data Science Club, Rice Ventures, Peer Academic Advisor, Freshman Student Representative
  • 🟢🟡 Duncan College 🦘 🟡🟢
  • Scholarships: James H Durbin Scholarship
E

Escuela ORT

2014 - 2019
High School Diploma, Information and Communication Technologies
  • Activities and Societies: Robotics Team, Biology Olympiad (National bronze 🥉), Junior Science Olympiad (2 x National Gold 🥇, 1 x International Bronze 🥉, Trainer's Assistant), Competitive Programming Olympiad (1 x National Silver 🥈, OII Participant, Trainer's Assistant)
  • Valedictorian / Argentinian Flag Bearer

Hackathons

  • x

    xAI Hackathon

    Palo Alto, CA, USA

    An enterprise ad portal for adversiers on X.com, powered by AI.
    Honorable Mention
  • T

    TreeHacks

    Stanford, CA, USA

    A fully end-to-end, AI-powered ad generator.
    Grand Prize WinnerDevpost
  • F

    Future House Hackathon

    San Francisco, CA, USA

    Platform for human-agent collaborative knowledge graph construction.
    Second Place
  • B

    Bio x ML Hackathon

    Cambridge, MA, USA

    Graph AI system that answers biomedical queries on knowledge graphs for precision medicine.
    Scale Medicine WinnerDevpost
  • T

    TreeHacks

    Stanford, CA, USA

    AI-enabled physician assistant for automated clinical summarization and question generation.
    Patient Safety Challenge WinnerDevpost
    •

Publications

  • Under Review

    KnowGuard: Knowledge-Driven Abstention for Multi-Round Clinical Reasoning

    Xilin Dang*, Kexin Chen*, Xiaorui Su, Ayush Noori, Iñaki Arango, Lucas Vittor, Xinyi Long, Yuyang Du, Marinka Zitnik, Pheng-Ann Heng

    ICLR 2026

    arXiv OpenReview

  • Graph AI generates neurological hypotheses validated in molecular, organoid, and clinical systems

    Ayush Noori, Joaquín Polonuer, Katharina Meyer, Bogdan Budnik, Shad Morton, Xinyuan Wang, Sumaiya Nazeen, Yingnan He, Iñaki Arango, Lucas Vittor, Matthew Woodworth, Richard C. Krolewski, Michelle M. Li, Ninning Liu, Tushar Kamath, Evan Macosko, Dylan Ritter, Jalwa Afroz, Alexander B. H. Henderson, Lorenz Studer, Samuel G. Rodriques, Andrew White, Noa Dagan, David A. Clifton, George M. Church, Sudeshna Das, Jenny M. Tam, Vikram Khurana, Marinka Zitnik

    arXiv preprint arXiv:2512.13724

    arXiv

  • A scalable platform to build the data layer of knowledge graph AI

    Lucas Vittor*, Iñaki Arango*, Ayush Noori*, Joaquin Polonuer*, Marinka Zitnik

    New Perspectives in Graph Machine Learning @ NeurIPS 2025 (NPGML)

    OpenReview

  • Enabling multi-agent collaboration in knowledge graph environments

    Iñaki Arango*, Ayush Noori*, Lucas Vittor*, Joaquin Polonuer*, Marinka Zitnik

    Scaling Environments for Agents Workshop @ NeurIPS 2025 (SEA)

    OpenReview

  • Prefix and Output Length-Aware Scheduling for Efficient Online LLM Inference

    Iñaki Arango*, Ayush Noori*, Yepeng Huang, Rana Shahout, Minlan Yu

    Sparsity in LLMs Workshop @ ICLR 2025 (SLLM)

    OpenReview

  • Multi-objective generative AI for designing novel brain-targeting small molecules

    Ayush Noori*, Iñaki Arango*, William E. Byrd, Nada Amin

    Generative and Experimental Perspectives for Biomolecular Design Workshop @ ICLR 2024 (GEM)

    The strict selectivity of the blood-brain barrier (BBB) represents one of the most formidable challenges to successful central nervous system (CNS) drug delivery. We use multi-objective generative AI to synthesize drug-like BBB-permeable small molecules with predicted binding affinity against dopamine receptor D2, the primary target for many clinically effective antipsychotic drugs.

    arXiv OpenReview

Selected Writings

See all posts →