Javier López-Contreras

Javier López-Contreras

A large-scale benchmark for few-shot program induction and synthesis
Ferran Alet*, Javier López-Contreras*, James Koppel, Maxwell Nye, Armando Solar-Lezama, Tomás Lozano-Pérez, Leslie Pack Kaelbling, Joshua B. Tenenbaum
International Conference in Machine Learning, 2021
website, github

We generate a large quantity of diverse real programs by running code instruction-by-instruction and obtain I/O pairs for 200k subprograms.

Research Experience
Undergraduate Thesis in Mathematics at U. C. Berkeley


Undergraduate Thesis supervised by Sug Woo Shin funded by Erasmus+, CFIS-Mobility and MOBINT Scholarships. On Artin's Conjecture about primes with a prescribed primitive root. Thesis, Slides.

Undergraduate Research in Mathematics at ICMAT

Summer 2022

Supervised by Daniel Macias and sponsored by a IntroSO Scholarship. The aim of the project was to read the classical literature of Number Fields and Elliptic Curves. Slides

Undergraduate Research in Machine Learning at MIT

2020-2021 and Summer 2021

Mentored by Ferran Alet and supervised by Leslie Pack Kaelbling and Tomas Lozano-Pérez at Learning & Intelligent Systems Group. We created a big and naturallistic Code Synthesis dataset. Full time during Summer 2021 and part time the rest of the university year.

Teaching Experience
Teaching Assistant / Mentor


Teaching Assistant for 1st year students of my double major program. Reviewing Linear Algebra, Affine Geometry, Single Varibale Calculus, Differential Calculus, Math Fundamentals, Discrete Math, Programming 1 and 2 and Introduction to Computers.

Software Engineering Experience
Google Software Engineer Internship

Summer 2020

Open source at Dart Language Team. Implemented a language feature via a compiler transformation profiting from the novel Dart Macro System.

Google Site Reliability STEP Internship

Summer 2019

In a team of Site Reliability development, working in Prodspec and Annealing, internal large scale automation tools and Autodash, their common interface. Revamped deprecated emergency protocols in Autodash.

Last Update: September 1st 2023