Skip to content
Navigation menu
Search
Powered by Algolia
Search
Log in
Create account
DEV Community
Close
#
mlops
Follow
Hide
Posts
Left menu
đź‘‹
Sign in
for the ability to sort posts by
relevant
,
latest
, or
top
.
Right menu
Never lose a training run again: a checkpoint-and-resume playbook for ephemeral GPUs
Tanay Joshi
Tanay Joshi
Tanay Joshi
Follow
Jun 23
Never lose a training run again: a checkpoint-and-resume playbook for ephemeral GPUs
#
machinelearning
#
python
#
mlops
#
learning
5
 reactions
Comments
1
 comment
6 min read
Stop building custom wrappers for your ML models.
Renato Marinho
Renato Marinho
Renato Marinho
Follow
Jun 24
Stop building custom wrappers for your ML models.
#
ai
#
mlops
#
python
#
productivity
Comments
Add Comment
4 min read
Using the channels-last memory format reduced the latency of our conversation backbone by 22%
Elise Moreau
Elise Moreau
Elise Moreau
Follow
Jun 24
Using the channels-last memory format reduced the latency of our conversation backbone by 22%
#
pytorch
#
computervision
#
machinelearning
#
mlops
1
 reaction
Comments
Add Comment
4 min read
Machine learning in production: the model is the easy part
Mridul Nagpal
Mridul Nagpal
Mridul Nagpal
Follow
Jun 24
Machine learning in production: the model is the easy part
#
ai
#
machinelearning
#
mlops
3
 reactions
Comments
1
 comment
3 min read
ML Observability on EKS: Logs, Metrics and Tracing Head-to-Head
Fernando Azevedo
Fernando Azevedo
Fernando Azevedo
Follow
Jun 23
ML Observability on EKS: Logs, Metrics and Tracing Head-to-Head
#
dataplatforms
#
eks
#
mlops
#
observability
Comments
Add Comment
11 min read
Benchmarking 5 LLM providers on one eval set, no SDK per vendor
Marcus Chen
Marcus Chen
Marcus Chen
Follow
Jun 23
Benchmarking 5 LLM providers on one eval set, no SDK per vendor
#
machinelearning
#
llm
#
mlops
#
devops
Comments
Add Comment
4 min read
Building a Self-Hosted MLOps Platform from Scratch with FastAPI, PostgreSQL, GCS, and Docker
SHIVAM UPADHYAY
SHIVAM UPADHYAY
SHIVAM UPADHYAY
Follow
Jun 23
Building a Self-Hosted MLOps Platform from Scratch with FastAPI, PostgreSQL, GCS, and Docker
#
ai
#
devops
#
python
#
mlops
Comments
Add Comment
4 min read
temperature=0 didn't make our LLM evals reproducible
Marcus Chen
Marcus Chen
Marcus Chen
Follow
Jun 23
temperature=0 didn't make our LLM evals reproducible
#
machinelearning
#
llm
#
mlops
#
infrastructure
Comments
Add Comment
4 min read
The SDXL VAE overflow that decoded black images in fp16
Elise Moreau
Elise Moreau
Elise Moreau
Follow
Jun 23
The SDXL VAE overflow that decoded black images in fp16
#
pytorch
#
computervision
#
machinelearning
#
mlops
1
 reaction
Comments
Add Comment
4 min read
Harvesting a regression test set from gateway logs with a plugin
Marcus Chen
Marcus Chen
Marcus Chen
Follow
Jun 22
Harvesting a regression test set from gateway logs with a plugin
#
mlops
#
llm
#
machinelearning
#
infrastructure
Comments
Add Comment
4 min read
Semantic caching our flaky-test summariser: 58% fewer LLM calls
claire nguyen
claire nguyen
claire nguyen
Follow
Jun 22
Semantic caching our flaky-test summariser: 58% fewer LLM calls
#
sre
#
devops
#
llm
#
mlops
Comments
Add Comment
4 min read
If a 270M Model Already Worked, Why Did I Fine-Tune a 7B One?
Suman Nath
Suman Nath
Suman Nath
Follow
Jun 21
If a 270M Model Already Worked, Why Did I Fine-Tune a 7B One?
#
machinelearning
#
llm
#
mlops
#
ai
Comments
Add Comment
3 min read
Data Contracts in Production: Stop Trusting Your Upstream Sources
Gabriel Henrique
Gabriel Henrique
Gabriel Henrique
Follow
Jun 20
Data Contracts in Production: Stop Trusting Your Upstream Sources
#
dataengineering
#
python
#
data
#
mlops
Comments
Add Comment
5 min read
Perplexity held flat after INT4. Task accuracy dropped 7 points.
Marcus Chen
Marcus Chen
Marcus Chen
Follow
Jun 19
Perplexity held flat after INT4. Task accuracy dropped 7 points.
#
machinelearning
#
llm
#
mlops
#
pytorch
Comments
Add Comment
4 min read
The seam our tiled upscaler left on every 4K product render
Elise Moreau
Elise Moreau
Elise Moreau
Follow
Jun 19
The seam our tiled upscaler left on every 4K product render
#
mlops
#
computervision
#
pytorch
#
machinelearning
Comments
Add Comment
4 min read
đź‘‹
Sign in
for the ability to sort posts by
relevant
,
latest
, or
top
.
We're a place where coders share, stay up-to-date and grow their careers.
Log in
Create account