Projects
A timeline of ML and data engineering work across retail, logistics, and operations.
Natural-language analytics over customer service conversations, built entirely in Snowflake. Cortex classifies each conversation into facets, a reviewed label map keeps the taxonomy canonical, and a semantic model lets business users ask questions in the Cortex Agent. Answers are aggregated-only by design for GDPR compliance.
Snowflake Cortex, Cortex Agents, MLflow, CloudWatch
Snowflake Cortex builds a persona profile per marketing segment; AWS Bedrock generates campaign copy for each one. All output is human-reviewed before go-live.
Snowflake Cortex, AWS Bedrock, AWS Lambda, Marketing Automation
Ingestion masks and stores Genesys conversations every five minutes; a state machine claims a batch of 100 or more and enriches it with Bedrock batch jobs, summarising and classifying it into the CRM wrap-up. A weekly evaluation pipeline samples the output and scores it with an LLM judge.
AWS Bedrock, SageMaker Pipelines, DynamoDB, Snowflake, MLflow, Python
Central MLflow tracking server on SageMaker with an S3 artifact store, provisioned via Terraform. A SageMaker Pipeline evaluation step compares each run against a hard threshold and a benchmark run; Alertmanager deduplicates and throttles before publishing to SNS for Teams.
Terraform, MLflow, S3, AMP, SNS, GitLab, SageMaker Pipelines
Freelance engagement under NDA for a procurement officer. A hierarchical taxonomy clusters highly specialized purchase order line items by semantic similarity, with hybrid BM25 and dense vector search plus cross-encoder reranking for benchmarking against price history.
Sentence Transformers, BERTopic, OpenAI API, ChromaDB, Streamlit
A monthly SageMaker Pipeline measures the causal retention effect of every completed A/B campaign, with automated checks for randomization, statistical power, and significance. Results surface in Tableau, replacing manual post-hoc analysis in spreadsheets.
SageMaker Pipelines, Snowflake, Tableau
Started as a HuggingFace agents course project. smolagents orchestrates scrapers across rental sites; a GitHub Actions cron job checks for new listings every morning and sends matches over WhatsApp.
smolagents, Selenium, Python, WhatsApp API, GitHub Actions
A single XGBoost churn model deployed across 5 countries via SageMaker and GitLab CI/CD, each with its own data warehouse and marketing system.
AWS SageMaker, GitLab CI/CD, Snowflake, XGBoost, MLflow
NLP pipeline maps public commodity and gas price indices to the raw material components in supplier contracts, measuring the lag and magnitude between index movements and supplier price requests.
Python, NLP, Snowflake, Public Gas & Commodities APIs
Natural-language analytics over customer service conversations, built entirely in Snowflake. Cortex classifies each conversation into facets, a reviewed label map keeps the taxonomy canonical, and a semantic model lets business users ask questions in the Cortex Agent. Answers are aggregated-only by design for GDPR compliance.
Snowflake Cortex, Cortex Agents, MLflow, CloudWatch
Snowflake Cortex builds a persona profile per marketing segment; AWS Bedrock generates campaign copy for each one. All output is human-reviewed before go-live.
Snowflake Cortex, AWS Bedrock, AWS Lambda, Marketing Automation
Ingestion masks and stores Genesys conversations every five minutes; a state machine claims a batch of 100 or more and enriches it with Bedrock batch jobs, summarising and classifying it into the CRM wrap-up. A weekly evaluation pipeline samples the output and scores it with an LLM judge.
AWS Bedrock, SageMaker Pipelines, DynamoDB, Snowflake, MLflow, Python
Central MLflow tracking server on SageMaker with an S3 artifact store, provisioned via Terraform. A SageMaker Pipeline evaluation step compares each run against a hard threshold and a benchmark run; Alertmanager deduplicates and throttles before publishing to SNS for Teams.
Terraform, MLflow, S3, AMP, SNS, GitLab, SageMaker Pipelines
A single XGBoost churn model deployed across 5 countries via SageMaker and GitLab CI/CD, each with its own data warehouse and marketing system.
AWS SageMaker, GitLab CI/CD, Snowflake, XGBoost, MLflow
NLP pipeline maps public commodity and gas price indices to the raw material components in supplier contracts, measuring the lag and magnitude between index movements and supplier price requests.
Python, NLP, Snowflake, Public Gas & Commodities APIs