Jan 2025 – Present · 1 yr 3 mos

• Building the first AI-native revenue engine for startups

Jan 2023 – Nov 2024 · 1 yr 10 mos · San Francisco Bay Area

• YC S24. Imagine if your pair programmer was the most advanced AI reasoning system out there, with knowledge of every line of code written on the web.
• We're on a mission to realize that vision. https://storia.ai/

Sep 2020 – May 2024 · 3 yrs 8 mos

• A consultancy helping companies across diverse verticals deliver machine learning and data-driven solutions to their hardest problems with special focus on NLP, recommendation systems, tabular data, and computer vision domains. We emphasize working on high-ROI projects with quick time to market. We've helped teams deliver 33% lift on key business metrics in <6 weeks.

Jan 2020 – Aug 2022 · 2 yrs 7 mos

• I created the premier educational platform to help people jumpstart their careers in machine learning and data science with 6000+ active users globally. Our resources are trusted by industry practitioners and students at the top institutions in the world including Stanford, Harvard, UC Berkeley, and more
• Confetti was acquired by Towards AI.

Mar 2019 – Jul 2021 · 2 yrs 4 mos

• I was a founding member of the Conversational Modeling team, a new group within Alexa whose charter was to combine the best of research and product on ambitious special projects for the future of the platform.
• Together with Dilek Hakkani-Tur, we grew the team from 2 to ~20 people. We tackled every part of the dialogue systems stack from NLU, NLG, and knowledge representation to multimodal conversational experiences.
• A few of the things I worked on while at Amazon:
• Designed our team's cloud infrastructure to enable rapid development and experimentation
• Architected and maintained our shared codebase including establishing best programmatic practices for agile collaboration
• Lead Alexa's first large-scale model training initiatives. I rolled my own infrastructure to train language understanding systems across 64+ GPUs on 1TB+ of conversational data. Distributed systems are fun.
• Built state-of-the-art machine learning models in NLU, NLG, and information retrieval
• Published at leading AI conferences including NeurIPS, AAAI, NAACL, SIGDial, etc.
• Incorporated our technology advances into the Alexa platform so that we could delight customers
• Ran the internal Alexa AI Deep Learning Blog which went out to 1600+ subscribers every week
• Managed 4 PhD researchers

Jan 2018 – Jan 2019 · 1 yr

• I developed iOS and Android apps for teaching artificial intelligence principles that have been used by 3000+ people around the world.
• iOS: https://apps.apple.com/us/app/artificial-intelligence-school/id1369987569
• Android: https://play.google.com/store/apps/details?id=com.theaischool

Nov 2017 – Mar 2019 · 1 yr 4 mos · San Francisco Bay Area

• I was the 8th hire at rideOS, an autonomous mobility startup that raised $34 million in venture capital from Sequoia Capital and Next47 and was acquired by Gopuff.
• As one of the early team members, I worked on our core routing solutions, built out monitoring technology for all of our infrastructure, and helped build our ridehailing and dispatch services.

Sep 2015 – Mar 2016 · 6 mos

• I was a teaching assistant for several computer science classes at Stanford, including CS 224N (Natural Language Processing) and CS 109 (Probability and Statistics for Computer Scientists). Responsibilities included grading assignments and exams, organizing and running study groups, and holding exam review sessions.

Nov 2014 – Dec 2017 · 3 yrs 1 mo

• I worked on various problems in task-oriented dialogue and natural language understanding broadly as part of the Stanford Natural Language Processing Group, including leading a multi-year collaboration between our lab and the Ford Research and Innovation Center. We pioneered many novel deep learning-based techniques for conversational systems that have been used extensively throughout the research community.
• My primary collaborators were Christopher Manning, Percy Liang, and Christopher Potts. For a look at some of our work, check out www.mihaileric.com

Jan 2014 – Jun 2014 · 5 mos

• Responsible for teaching section for Stanford's introductory computer science courses.
• Responsible for grading assignments and exams, offering assignment helper hours, and
• teaching weekly sessions on course-related material.

Jun 2013 – Aug 2013 · 2 mos

• Worked in lab of Dr. Subhasish Mitra under Max Shulaker. Studied the use of directed self-assembly of block copolymers to improve carbon nanotube density and alignment.

Jun 2015 – Sep 2015 · 3 mos · San Francisco Bay Area

Jun 2014 – Sep 2014 · 3 mos · New York City Metropolitan Area

Jan 2012 – Sep 2013 · 1 yr 8 mos

• Responsible for answering student questions during classes as well as on the classroom message board. Responsible for posting supplementary problems for students. Courses taught: Mathcounts Problem Series, Introduction to Algebra, Introduction to Programming.

Jan 2011 – Sep 2013 · 2 yrs 8 mos

• Responsibilities include grading mathematics problem sets and examinations for various courses such as Introduction to Counting and Probability, Introduction to Geometry, Algebra 1-3, Precalculus, and Calculus. Help students with understanding mathematics concepts through detailed explanations and comments.

Jun 2011 – Aug 2011 · 2 mos

• Worked in lab of Dr. Mahdi Sanati. Responsibilities included conducting first principle calculations on intermetallic alloy systems, operating full-potential (FP-LAPW) and pseudopotential (VASP) programs, and interpreting data from these programs.

Jun 2010 – Aug 2010 · 2 mos

• Worked in lab of Dr. Miroslav Peric. Responsibilities included preparing chemical solutions and gathering data via the use of an electron paramagnetic resonance spectrometer.