Hi I’m Jeanne,
I am a computer and electrical engineering doctorate with advanced degrees from the University of McGill and a technology oriented creative from Montreal, Canada.
Skills & knowledge:
Javascript, Python, Java, C, Solidity, Typescript
Express, React, React Native, Flask, Node.js, Next.js
Linux, Ubuntu, Postman, AWS, POSTGRESQL, Knex, Vercel, Gitlab, Github, Jira, Confluence, VSCode, Google Collab, Jupyter notebooks.
GPU Cloud services: Runpod.io, Vast.ai, Google Cloud.
Claude, Mixtral, Llama, Gemini, GPT-4, GPT-3, chatgpt, Langchain, Stable Diffusion, Hugging Face, Alpaca Lama.
Ethers.js, Web3.js, Infura, Alchemy, Hardhat, OpenSea, Zora, Sound, Catalog, Royal, 0xSplits, Arweave/Bundlr, Web3Storage, NFTStorage, IPFS/Filecoin, Graphql, Airstack, Walletconnect, Livepeer, XMTP Protocol.
ERC-20, ERC-1155, ERC-721, ERC-6551.
ETH, Polygon, Arbitrum, Optimism, Goerli, Sepolia.
Quantum computing, quantum machine learning.
Formal training:
B.ENG. in Computer and electrical engineering with a focus on Advanced Communication Systems
M.ENG. in Solar Cell heterojunction devices
Ph.D. in Flexible Ultrasonic Transducer Fabrication
Twice awarded for Best Paper by the Institute of Electrical and Electronics Engineers (IEEE) and by the Institute of Physics (IoP), I worked at the National Research Council of Canada for my doctoral thesis.
I am a research fellow with Seeds of Science, an independent research collective.
My tiles:
https://gitlab.com/jeannebeem
https://github.com/jeannemtl
About me
An early adopter of bitcoin circa 2011 (OG pc miner) with experience in intellectual property and exposure to several aspects of building. I’m passionate about crypto, ai, quantum, tech startup dynamics, new social trends and making interesting products. With a strong sense of community, I work in web3 as a full stack engineer and provide my skills & unique knowledge as an engineering consultant.
Skilled in:
Ingesting data on deep technology and complex information.
Performing R&D on technical trends, anticipating & communicating them.
Productizing upon the R&D, proposing and designing new features into existing infrastructure.
Writing backend API service endpoints.
Reading smart contracts and writing scripts to access on chain data.
Prompting LLM’s to do research, debug code, to find dependency conflicts, to read other people’s code bases and crafting targeted lists of instructions to execute at scale, in a self-managerial way.
Discerning high impact talent in creative people, leaders & assembling teams.
Quantum:
I keep a pulse on topics related to quantum computing & quantum machine learning and how it intersects with artificial intelligence. I like to follow progress in the industry’s quantum roadmap and generally understand where things need to go to achieve an industry breakthrough. I also keep abreast on quantum related startups and their progress. Most of my understanding in these fields was derived while I was part of the first cohort on Quantum Machine Learning at the Creative Destruction Lab in Toronto, Canada.
Web3:
I productised and integrated web3 features and backend API endpoints for beem.xyz, a video streaming platform for independent creators.
Work:
The implementation of decentralised storage of videos uploaded to Arweave using Bundlr and IPFS pinning with web3Storage and NFTStorage with backend endpoints to the Postgresql database.
The implementation of 0xSplits revenue splitting through split contract creation and the implementation of CREATE2, based on a typescript backend and javascript frontend using redux.
The implementation of GraphQL queries and Airstack API’s with Alchemy, Ethers.js for populating NFT collections, getting token data.
The migration of Walletconnect v1 to v2.
The design for a custom smart contract to act as an interface between users and a suite of web3 features for the benefit of the former who may not be native to web3.
I also worked for a web3 social music sharing app. My task was to build out API endpoints to periodically ingest on-chain music NFT data and to productize this ingestion as part of their existing internal API. I wrote scripts to read smart contract on chain data.
Intellectual property and fintech finance:
I worked as a patent engineer delving into intellectual property topics having worked at a boutique law firm in Montreal, gaining invaluable understanding about drafting patents, the application process and the examination process in Canadian, American and International jurisdictions and certain legal aspects of patent rules and regulations.
I have also worked for a neo-bank fintech start up based out of Paris, France, investigating on their behalf its potential as a money-service-business or chartered bank to operate in Canadian territory.
NFT’s and digital art:
Artist Martin Lukas Ostachowski & I organized the first NFT conference in Montreal in 2022 aimed at onboarding local Canadian artists to the crypto & NFT space & fostering open minded discussions. Check out what we accomplished in just a few months entirely from generous donations by the crypto community @MintMtl !
Artificial intelligence:
I have been using GPT-3 & GPT-4 with early access and chatGPT since it’s release. With a good understanding of the concept of vector embeddings, I take an interest in the research field on how to manipulate them for building out AI based tools.
Over the years, I have also connected with talented engineers & creatives and built out some prototypes with them, so please check out some of the projects I’ve been involved with below!
Orchid - 2024
A model merging engine built to select models in a timely manner that is appropriate for every user.
Orchid understands when models can be merged and when hybrids can be used. In this way, it keeps generating new hybrids while serving users with a sophisticated knowledge analysis tool.
ETH GLOBAL - WATERLOO SUBMISSION - 2023
My submitted project was an implementation of LLM’s and ERC-6551.
The project ingests a prompt that is presumed to be relevant to two pieces of embedded text though it could be more. I embeded a book by Nietzche and Jane Austen’s Pride & Prejudice as an example. The prompt was embedded as well. An embedding library was used to perform a nearest neighbour search between the prompt embedding and each of the two embedded pieces of text.
The result is a new vector embedding meant to represent the three way relationship which could possibly be used to fine tune other LLM’s or the original model. This new embedding representing the relationship is then stored as metadata on an NFT that by virtue of the being an ERC-6551 NFT has its own wallet. Having its own wallet means the NFT, which holds the embedding vectors representing the relationship between three pieces of text, is now able to autonomously transact itself in a hypothetical marketplace of AI text embeddings.
As they are stored as metadata on NFT’s, the embeddings may also be transacted by humans as valuable items for the refinement of personal LLM’s, thus combining a decentralized blockchain protocol, the ERC-6551 and LLM’s, within the context of an autonomous or voluntary marketplace.
In other words, it’s designed purpose is to improve LLM’s as cognitive tools for both humans and AI agents through an NFT marketplace of relationship bearing vector embeddings stored as metadata on token bound NFTs.
Training voice models - 2023
I trained my own voice model using the open source library called so vitz svc fork and recordings of my own voice, the results of which can be shown here where I do inferencing on short clips of videos and focused on expressibility of the voice.
Here I am as Uma Thurman, Quentin Tarantino, Kazuo Ishiguro. And finally, from a clip of the Richard Linklater movie, Waking Life.
Contextualize - 2020
A context-based recommendation engine built on GPT-3, OPEN AI’s then most advanced natural language model.
I built this to call the GPT-3 API in 3 different ways. The aim was to give context to a topic by training GPT-3 to think and speak like three different persons.
This was accepted for demonstration as one of the first GPT-3 demos presented to Pioneer.app in 2020 as part of “The First Open GPT-3 Demo Day”. The prototype was submitted as a “contextual reference recommendation engine using gpt-3” meant to serve as a research tool.
Bumpin:
Quantum dating - 2017
Bumpin was a 2017-2018 AI-based dating app, meant to scrape users' social media metadata and images to generate representations of people using higher representation vectors, calculating metrics between users and matching them based on this evaluation as a way to match individuals on a different plane.
This was a project proposed as part of the Creative Destruction Lab’s Quantum Machine Learning stream.
Cocomint: An AI engine - 2019
Cocomint was an AI engine for fashion discovery built with machine learning APIs to find similar outfit items based on a single snapshot.
Quaintance - 2018
Quaintance was an AI engine for IKEA showroom discovery to purchase furniture according to a person’s overall style & tastes.
These are the office and bedroom results from the prototype AI trained on IKEA room layouts, populating the most “similar” IKEA room layouts from an input image.
The S.W.A.P. Team
The S.W.A.P. Team was a community driven non-profit I volunteered at for over five years, acting as an operations VP helping to organize city-wide clothing swap pick ups and events. Entirely driven through volunteer work, items from the events which were not swapped were always donated away. Its mission was to divert textiles from landfills and give back to the community through events and donations. Aleece Germano, the founder created it as a turn-key service, entirely sustained by volunteers during the initial rise of the sharing economy. I helped launch a Kickstarter campaign in collaboration with Montreal Couture to fund activities of the clothing swap.
Under the banner “Take off your clothes” we helped community leaders organize several clothing pick ups and city-wide clothing swap events for one-to-one item exchanges in over 10 cities in North America including NYC, Montreal, Toronto and Calgary. I oversaw the logistics and organization. As a sample, the first event in Calgary received 196 participants, saw 4147 items swapped and 1060 items donated to the Salvation Army. This was a half-day event.
Vlyy video -2014
Vlyy video was a short video clip application meant as a new social media platform adapted for the rise of VR and AR. Around the time of Snapchat and Vine, it was built to be a short clip video app, advertised as a “visual timeline” where one could take daily clips of video and they would be organized such that they could be visually recalled on a wide web view. The web view was built for maximum visual design, by having consecutive scenes in a full view and searchable as groups of videos as an intuitive visual space.
