About Vooma
Enterprise software is in the midst of a revolution. The world is shifting from "systems of record" to "systems of automation" - LLM powered agents and orchestrators that perform complex business tasks on behalf of and in concert with humans, with increasing levels of autonomy.
Vooma's mission is to build the AI orchestration platform for America's $800+ BN trucking industry. Everything we touch in the physical world is moved on a truck - it's the backbone of the US economy. Yet logistics today is still extremely manual, which makes it one of the hungriest industries for the shift toward AI Agents.
At Vooma, the products you'll build will help shippers, freight brokers, and trucking companies win more business and seamlessly execute freight. Our AI agents operate across voice, email, sms (and more) to automate critical workflows across the load lifecycle.
Vooma is backed by top-tier investors including Index Ventures, Craft Ventures, Y-Combinator, and CEOs, founders and executives from the logistics industry.
About the role
Your role, should you choose to join us, will be an AI Engineer on our founding team.
You're the right person for this role if you're excited to work at the frontier of applied AI, training and iterating advanced models that work in the real world. You’re driven by turning cutting-edge capabilities into reliable, high-performance systems that deliver highly scaled impact.
You're a great fit if you're known for going deep on model behavior and building systems that continuously improve through well-designed data flywheels. You find energy in shipping fast, learning from real usage, and iterating toward systems that feel almost magical in their effectiveness.
Your responsibilities will include:
Training, fine-tuning, and deploying state-of-the-art multimodal models across a range of real-world tasks
Designing and implementing evaluation frameworks to rigorously measure model performance and guide iteration
Building and scaling data flywheels - collecting, curating, and generating high-quality datasets to continuously improve model outcomes
Developing systems for live learning, feedback incorporation, and continuous model adaptation in production
Implementing techniques like negative mining to harden models against edge cases and failure modes
Owning the full lifecycle from experimentation → validation → deployment → monitoring
Collaborating closely across engineering and product to integrate models into reliable, high-performance systems
Must haves:
Hands-on experience working with modern foundation models (LLMs, multimodal models, or similar) in production settings
Strong intuition for model behavior, evaluation, and failure modes
Experience with fine-tuning, training pipelines, and dataset construction
Familiarity with techniques like RLHF, synthetic data generation, or active learning (not required, but highly relevant)
Comfort working across Python-based ML stacks and Typescript-based production systems
A bias toward action - you run experiments, measure results, and iterate quickly
The mindset of an owner: you care about outcomes, not just outputs, and push systems to actually work in the real world
Why You'll Love Working Here
Rare opportunity to join a formidable team building an enduring company
Direct impact - your work will be actively helping to food get on the table for millions of households
Cutting-edge technology - work with the latest in agentic AI using modern stacks (Next.js, GraphQL, Node, OpenAI, Anthropic)
Competitive compensation with significant equity upside potential
Comprehensive benefits including medical, dental, and vision coverage
In-person collaboration in San Francisco with a talented, experienced founding team
This role is based in person in San Francisco, CA (not remote).