You will get an end-to-end MLOps pipeline in AWS/GCP
Project details
I don't just write code — I build data infrastructure that your team can actually own, maintain, and scale. Every engagement I take on is production-ready by default: tested, documented, governed, and handed over with a walkthrough so nothing is a black box.
What sets me apart is the combination of depth and breadth. I've worked across the full modern data stack — Snowflake, dbt, Airflow, BigQuery, SageMaker, Looker — at startups moving fast and enterprises with complex compliance requirements. I've led platform overhauls, cloud migrations, and MLOps builds end to end, not just contributed to them.
I also bring a consulting mindset. Before writing a single line of code I make sure I understand your business problem, not just your technical requirements. That means fewer surprises, less rework, and a final product that actually solves the right problem.
Simply put — you get a senior engineer who treats your project like it's their own.
What sets me apart is the combination of depth and breadth. I've worked across the full modern data stack — Snowflake, dbt, Airflow, BigQuery, SageMaker, Looker — at startups moving fast and enterprises with complex compliance requirements. I've led platform overhauls, cloud migrations, and MLOps builds end to end, not just contributed to them.
I also bring a consulting mindset. Before writing a single line of code I make sure I understand your business problem, not just your technical requirements. That means fewer surprises, less rework, and a final product that actually solves the right problem.
Simply put — you get a senior engineer who treats your project like it's their own.
AI Development Type
Deep Learning, Knowledge Representation, Model Tuning, Recommendation System, Software MaintenanceAI Tools
Amazon SageMaker, Apache MXNet, Google AutoML, Keras, MATLAB, MLflow, OpenCV, PyBrain, PyTorch, TensorFlowAI Development Language
PythonWhat's included
| Service Tiers |
Starter
$350
|
Standard
$900
|
Advanced
$2,200
|
|---|---|---|---|
| Delivery Time | 5 days | 14 days | 21 days |
Number of Revisions | 1 | 2 | 3 |
AI Model Integration | |||
Detailed Code Comments | - | - | |
Knowledge Graph | - | ||
Model Documentation | - | ||
Ontology | - | ||
Source Code | |||
Taxonomy | - | - |
About Umang
Data Engineering & MLOps | Data Warehouse, Pipelines, BI & ML Infra.
Westmead, Australia - 7:37 am local time
My stack: Redshift/Databricks/BigQuery/Snowflake · DBT · Airflow/Dagster · Python · SQL · Terraform/CloudFormation. I've replaced broken Redshift stacks with modern Snowflake warehouses, migrated pipelines to GCP from scratch, and delivered ML-ready data infrastructure at companies like Mosh, Sonder, and Westpac.
I bring technical depth and clear communication — I can sit with your engineering team in the morning and present architecture to your leadership in the afternoon. I care about doing it right: tested, documented, governed, and maintainable.
If you need a reliable data engineer to design, build, or fix your data platform — let's talk.
Steps for completing your project
After purchasing the project, send requirements so Umang can start the project.
Delivery time starts when Umang receives requirements from you.
Umang works on your project following the steps below.
Revisions may occur after the delivery date.
Discovery & Assessment
Review your existing models, deployment process, data sources, and infrastructure. Understand your retraining frequency, latency requirements, and team's technical capabilities.
Architecture Design
Propose the end-to-end MLOps architecture covering feature store, training pipeline, model registry, deployment strategy, and monitoring approach. Shared for your sign-off before build.