Building AI agents, Global Food Systems Challenge, PCB boards, Stanford, and more fun AI projects
Vinaya Sharma | November and December Newsletter ⛄️🧬
Hey! If you’re new here, I’m Vinaya and welcome to Bits and Bytes! I'm a 17 y/o tech nerd from Toronto that’s always building. 👩💻
In this monthly newsletter, I share Bits and Bytes of my learnings to help you grow from my key insights! 🌱 Stay tuned for bits of self-improvement techniques and bytes of groundbreaking tech innovations.
I decided to record a quick update for my newsletter this month. Check it out and feel free to learn more by reading and clicking the links below! 👇
LLMs for Drug Discovery 💊
Continuing with my drug discovery rabbit hole, I wrote an article on using LLMs for drug discovery. I dive into the Regression Transformer which engineers sentences with chemical information and SMILES tokens and makes the model guess the missing fields.
This deep dive plus the other mini bioinformatics projects I completed by following the Data Professor made me recognize how big of an issue the data is.
We have these huge datasets we need to sort through to build ML models and it’s even worse for all the wet lab scientists, bioinformaticians, and computational biologists who are trying to make sense of and understand the data. Drug properties, structures, and interactions are very important to understand to design good drugs and to enable personalized medicine, genomics research, and drug repurposing. In recognizing this gap in mundane data tasks, I built BioBytes.
Building an AI agent for biochemistry-related EDA 📊
The goal with BioBytes was to build a super simple platform that lets you talk to your data with natural language to understand the statistical makeup, quickly summarize and get you up to date with what data is being shown, automatically turn data into visuals, spots interesting patterns, and clean up data to run machine learning models.
I got to present this project at the TKS Activate showcase and got some more feedback and validation on the tool so far.
You can learn more about the tool in the project memo, and here’s a run-through of the 5-minute Activate presentation.
Other exciting things that happened this month 🪅
Open sourcing the one million dollar food innovation challenge data with the Seeding Future Foundation and IFT.
Building a water level detection system. I’ve been slowly taking this from an Arduino to a Perfboard, and now onto my own PCB.
Got into Stanford!
Upcoming plans for the following months 👷♀️
In these upcoming months, I’ll be focusing on more Bioinformatics tasks I can automate with BioBytes. I’m trying to turn BioBytes into a tool that assists Bioinformaticians in the daily tasks that they pursue. Right now I’m trying to understand what data they commonly work with, how big that data is, and what kinds of analysis they run on this data to then hopefully build robust AI agents that can speed up processes for them.
I’m also leading the development of an AI bridge fault detection project, and as always, I’m tinkering with other AI projects in my free time - this time a fun one with Raspberry Pis.
Love this Vinaya! Really excited to see how BioBytes progresses.
upwards and onwards