Yet another FDE in the Bay... Engineering @FurtherAICom.
Financial District
- Tweets 192
- Following 155
- Followers 39
- Likes 31
open envs, evals, and training for code rl. rare to get all three in one drop
Super happy to release SmolDataEnvs: 5,000 verifiable RL environment tasks for hill-climbing small models in code and data science by @adithya_s_k
100% open source: environments, evals, training!
huggingface.co/datasets/Fine…
1
37
Thanks for the donuts @infragrid
Who wants free donuts?
Going around to startups this week dropping off donuts. Reply if your team wants some
1
2
7
371
the best part of the FDE role is increasingly not the setup work. it’s the feedback loop.
at FurtherAI, that means partnering with customers to build and iterate evals, understand the costly edge cases, and making each workflow more reliable over time.
full piece below.
and we’re hiring across engineering, GTM, and more!
Software now does the work that used to define the Forward Deployed Engineer role.
An FDE used to sit with a customer and turn a model into something that ran in production: writing the prompts, choosing the models, building the integrations, creating an eval set, testing until the workflow held up. Our platform does that now.
So, what does that mean for the future of the role?
It moves to what the machine cannot settle. Someone still has to decide what "correct" means, and in insurance, not every error costs the same. Judging which errors a workflow can absorb, and when it's ready for production, stays human.
The rest of the job is building what the platform can't do yet, so the next insurer gets it as standard.
Our Co-Founder and CEO @amangour30 wrote about where the role goes from here. Full piece linked below:
15
343