steveblackmon

Steve Blackmon

@steveblackmon
GitHub Profile
collaborative and process-focused
Steve is a pragmatic and experienced reviewer who focuses on workflow optimization, proper deployment practices, and architectural consistency. He tends to approve PRs readily while providing constructive guidance on process improvements and code organization.
101
Comments
84
PRs
9
Repos
28
Avg Chars
2
Harshness

Personality

Process-oriented Pragmatic Workflow-conscious Supportive of testing Architecture-focused Deployment-aware Detail-oriented about naming conventions Forward-thinking about automation

Greatest Hits

"LGTM"
"can we merge this to develop and test it there first?"
"if we're going to change this class anyway, we should"
"lets add a unit test to check"
"Should this merge to develop first and then to main? One way flow of changes should be the goal"
"lets open a new PR so we aren't unable to merge"

Focus Areas

Common Phrases

"LGTM" "lets" "should" "first" "this" "change" "merge" "develop" "catalog" "check" "going to" "anyway" "can we" "may want to" "should this merge to develop first"

Sentiment Breakdown

positive
7
neutral
16
questioning
1
constructive
1

Review Outcomes

APPROVED
79
CHANGES_REQUESTED
4
COMMENTED
3

Most Reviewed Authors

dheerajchand
39
taha-tf
32
gabgoesfish
20
hongjingzhou
4
steveblackmon
4
jzachr
1
paveldudka
1

AI Persona Prompt

You are Steve Blackmon, an experienced code reviewer who focuses heavily on proper deployment workflows and process optimization. Your reviews are generally approving (you say 'LGTM' frequently) but you're always thinking about the bigger picture - deployment processes, branch management, and long-term maintainability. You frequently ask 'can we merge this to develop first?' and emphasize proper branching workflows. You're supportive of testing ('Thanks for making tests', 'lets add a unit test') and care about configuration management and naming conventions. You often suggest process improvements and think ahead about automation ('once we deploy X these routine deploy PRs wont look like this'). Use casual language with phrases like 'lets' instead of 'let's', and often frame suggestions as collaborative questions starting with 'can we' or 'should we'. You're practical about technical debt but always push for proper processes. When you see opportunities for improvement, you phrase them constructively. You appreciate when people follow best practices and aren't afraid to point out when AWS documentation should be followed for naming conventions. Keep your comments concise but actionable, and always consider the deployment and workflow implications of changes.

Recent Comments (25 total)

agentql-apps/#584 Add Databricks cache for Google Hotel ML mode [view]
On first read-through, LGTM. Thanks for making tests
agentql-apps/#542 Migrate to stream arns instead of names: Databricks kinesis streams migration · apps/google_hotel_sqs/app/common/kinesis_utils.py [view]
if we're going to change this class anyway, we should change this argument name from StreamName to StreamARN per https://boto3.amazonaws.com/v1/documentation/api/latest/reference/services/kinesis/client/put_records.html
agentql-apps/#542 Migrate to stream arns instead of names: Databricks kinesis streams migration · apps/google_hotel_sqs/app/common/kinesis_utils.py [view]
if we're going to change this class anyway, we should change this argument name from StreamName to StreamARN per https://boto3.amazonaws.com/v1/documentation/api/latest/reference/services/kinesis/client/describe_stream.html
tf-databricks/#92 Fix catalog detection (hive_metastore issue) and ModuleNotFoundError in jobs [view]
can we merge this to develop and test it there first?
tf-databricks/#92 Fix catalog detection (hive_metastore issue) and ModuleNotFoundError in jobs · .github/workflows/CD_develop.yml [view]
dab deploys to dev are competing for ownership of tables with workspace pipelines, i'd like to just do dab to staging and prod
tf-databricks/#116 Parameterize optimized_scrapes notebook and add base_parameters · resources/jobs.yml [view]
@gabgoesfish this seems important to fix, other lgtm
tf-databricks/#88 promote develop to main [view]
once we deploy https://github.com/tinyfish-io/github-control/pull/1095 these routine deploy PRs wont look like this
tf-databricks/#82 Refactor hotel_helpers.py into separate modules [view]
LGTM, may want to add static method decorators as advised by coderabbit
tf-databricks/#77 fix: Update demo notebook setup code and fix volume_config maxBytes issue [view]
Should this merge to develop first and then to main? One way flow of changes should be the goal
tf-databricks/#62 docs: Reorganize repo structure and add CI/CD documentation [view]
super nit but can we not all-upper-case the names of python source files? documentation upper-case names are fine
tf-databricks/#40 Add Unity Catalog volume-based configuration system (ML-1047) [view]
Is it possible for catalog to be an optional param and when not set the library uses a unity sdk trick to find out what catalog the current pipeline is in? The whole point of this is for the same code to work in all workspaces, and we had been expecting to use the config system to pass in a value for 'catalog'
tf-databricks/#27 Restore and update annotation system with new config system (ML-997) · src/notebooks/annotation_processing.py [view]
is this still accurate?
tf-databricks/#26 Add gh-initial-state pipeline for optimized scrape scheduling · src/pipelines/gh-initial-state/load_data.py [view]
i think its ok to omit {catalog}.{schema} that will use the current execution context of the pipeline
tf-databricks/#26 Add gh-initial-state pipeline for optimized scrape scheduling [view]
please modify to make the upstream branch 'develop' or re-open as new PR
tf-databricks/#29 confirmed pipeline end-to-end execution in development workspace [view]
<img width="1424" height="232" alt="image" src="https://github.com/user-attachments/assets/8b8d600c-7cd5-4805-b282-95295e991744" />