Exploring My Next Chapter in Machine Learning
Where I Am Today
In 285 days, my current role as a data analyst will come to an end. This is both a closing and an opportunity. I have worked as a data analyst for the last eight years, and it has been an incredibly rewarding experience. Working at all levels of the pipeline (including low-tech and no-tech environments), I’ve navigated constructive technical conversations in the context of complex people and power dynamics and financial constraints. I’ve analyzed hundreds of datasets, designed surveys, created visualizations, answered questions about data collection, met compliance requirements, manually entered and validated data, and built and improved R, Python, and SQL pipelines — all with awesome people. I will greatly miss and always be grateful for that experience.
I’m also excited. I feel ready to begin my professional machine learning journey. I have worked my way from being on the outside, to being a fast.ai student, a contributor and collaborator, and most recently, an open source maintainer. I want to bring that experience with me and create new ones with people working on fascinating ML problems.
What I’ve Been Building
I’ve hit three of my five goals for the year. I published 50 ML blog posts, 50 ML videos, and contributed to open source (as the maintainer of the stanford-futuredata/ColBERT library).
I have given over a dozen paper presentations in fastai study groups. I have spent dozens of hours working on independent research with other fastai community members.
I took the SolveIt course (AnswerAI), and the AI Evals course (Hamel Husain and Shreya Shankar).
I’m slowly building my online platform with 1800 followers across X and YouTube where I regularly post technical ML content.
I practice my values of resilience, consistency, thoughtfulness, and building publicly when publishing content.
Four Things I’ve Learned Along the Way
Do hard things while following your excitement and interest. I’ve found that excitement and interest are two strong signals my intuition gives me—“hey look into this, it’s important. I can’t explain why until I understand it.” It’s what led me to deeply studying the ColBERT papers, which led me to my role as a maintainer.
Balance commitment and prioritization. I work on it like I’m going to finish it, but deprioritize it when needed.
I’m always looking to publish. Every new problem I solve or concept I learn is an opportunity to write about it and understand it better.
Consistency doesn’t mean constancy. How I show up to my ML work and communities changes as situations in life change. What doesn’t change is showing up.
What I’m Looking For in My Next Role
I want work that’s exciting and keeps me learning. The right role gives me space to grow and the stability to go deep on real problems.
I care about efficiency. I don’t believe in throwing money at problems. I like getting to the root of things — especially when two things should match but don’t. Figuring out why is the fun part.
Teaching is core to how I work. I’ve taught STEM in high school and engineering at community college. For me, writing and explaining are how I learn. That’s why I blog, record videos, and work on open source.
The means and the end are the same. Solving the problem is itself the goal. And I’d rather do it with others than alone.
You have to let me look at data. That’s the single greatest skill I’ve developed as a data analyst.
Open source matters to you. I want to work with people who value my open source work — study groups, research, ColBERT maintenance — and see it as something to support and invest in.
I’ve been remote for eight years, and I plan to stay that way. I do my best with autonomy — following paths that might not look obvious at first. At the same time, I’m coachable. If someone I trust gives me advice, I take it seriously and put it into practice.
If you’re working on something exciting in ML and think I could contribute as an employee or consultant, I’d love to chat: vdbakshi at gmail dot com.
Looking Ahead
Over the coming months I’ll be continuing to write, publish, and share what I learn. I’m looking forward to connecting with others who are building the future of ML.