A platform that automates continual learning to let non‑technical users fine‑tune open‑source AI models with feedback loops
- Python
- PyTorch
- Kubernetes
- AWS
bottleneck › Scaling the continual‑learning pipeline and managing feedback data efficiently
contact
I can rebuild your feedback ingestion layer with Next.js to give you a fast, developer‑friendly UI and add a vector‑semantic search layer to instantly surface the most relevant training examples. This will cut your data‑pipeline latency and let your engineers focus on model improvements instead of plumbing.
email draft › no published address
intent: build for them (a concrete bottleneck on a stack you can work with)
Trajectory: a thought on Scaling the continual‑learning pipeline and managing feedbac
Hi Ronak, I saw you're hiring (Member of Technical Staff - Research, Member of Technical Staff - Design Eng) (https://jobs.ashbyhq.com/trajectory). I'd like to help Trajectory reach more of the teams it is built for. I came across Trajectory: A platform that automates continual learning to let non‑technical users fine‑tune open‑source AI models with feedback loops. One thing I noticed: scaling the continual‑learning pipeline and managing feedback data efficiently. I can rebuild your feedback ingestion layer with Next.js to give you a fast, developer‑friendly UI and add a vector‑semantic search layer to instantly surface the most relevant training examples. Would a short trial project to tackle this be useful? If this isn't relevant, just reply "no" and I won't follow up.
try it first
signals
- funding rss · Nov 2
- hiring Member of Technical Staff - Research, Member of Technical Staff - Design Eng, Member of Technical Staff - Generalist, Member of Business Staff ashby · Oct 6
qualification
- fit10000%
- reach5000%
- evidence8500%
timeline
similar companies
- why now › Newly found at series a stage
- why now › Raised money
- why now › Hiring (39d ago): Member of Technical Staff - Research, Member of Technical Staff - Design Eng, Member of Technical Staff - Generalist
score › fit 25/25 · identity 12/15 · people 20/20 · connections 15/20 · contact 8/20
missing › location · email · phone
people
- Ronak Malde · Co‑founder x
“Trajectory co‑founder Ronak Malde compared foundational models to a very smart PhD student”
- Michael Elabd · Co‑founder in
“co‑founder Michael Elabd argues that keeping humans involved in the training process makes Trajectory’s approach safer”
- Arjun Karanam · Co‑founder in
“Trajectory’s third co‑founder, Arjun Karanam, previously worked on generative foundation models at Apple”
connections
- Ronak Malde founded Trajectory
“Trajectory co‑founder Ronak Malde”
- Michael Elabd cofounded Trajectory
“co‑founder Michael Elabd”
- Arjun Karanam cofounded Trajectory
“Trajectory’s third co‑founder, Arjun Karanam”
- Sequoia invested in Trajectory
“led by Sequoia”
- Nvidia invested in Trajectory
“participation from Nvidia”
- Bessemer Venture Partners invested in Trajectory
“participation from ... Bessemer Venture Partners”
- Bain Capital Ventures invested in Trajectory
“participation from ... Bain Capital Ventures”