Praveen Radhakrishnan

AI Engineer · Data & AI Consultant

pavirad@gmail.com · linkedin.com/in/pavirad · github.com/0xtreme

Melbourne, Australia

AI engineer working on production GenAI, across retrieval, evaluation and guardrails. Twenty years of enterprise data platform work behind it, in banking, energy, superannuation and government. The retrieval layer is not somebody else's problem.

Selected AI work

  • Built an evaluation framework measuring faithfulness, answer relevancy and contextual precision, with evaluation datasets for ongoing model monitoring.
  • Designed the guardrail and observability layer for a production LLM system, covering content filtering, API latency and token usage.
  • Took a multimodal Azure OpenAI system from first design to production inside a government agency.
  • Twenty years building the data platforms underneath, from Databricks lakehouse and change data capture to warehouse migrations and ETL across regulated industries.

Experience

Frontier model evaluationAlignerr

Current
  • A/B prompt testing across frontier models, scoring reasoning and behaviour against detailed rubrics.
  • Reviewing scored evaluations for correctness and consistency of application.

Senior Consultant, Data & AI ArchitectSISU Solutions

Dec 2024 – May 2026

AI platforms for government child welfare agencies

  • Architected a multimodal Azure OpenAI case note summarisation system end to end, design through production. Text, audio and image through blob storage.
  • Implemented the evaluation framework measuring faithfulness, answer relevancy and contextual precision, with evaluation datasets and a metrics framework for model performance monitoring.
  • Designed the monitoring and observability layer tracking API latency, token usage and content filtering hits.
  • Architected a live call centre intelligence platform using streaming transcription on Azure Communication Services, with event driven processing via SignalR and Event Grid under 500ms, a vector embeddings pipeline for RAG and a Neo4j graph for network analysis.
  • Embedded with the government engineering team, working within enterprise security constraints, legacy data infrastructure and compliance requirements.

Senior Consultant, Data EngineerMUFG Pension & Market Services

Nov 2023 – Jun 2024

Australia's largest superannuation administration platform

  • Cut a critical SQL Server batch from four hours to 45 minutes.
  • Built a data quality framework catching anomalies before production, and automated reconciliation across 100M+ monthly transactions.
  • Built Power BI pipelines for APRA regulatory reporting, and incremental loading patterns to cut transfer volumes.

Senior Consultant, Data EngineerEnergyAustralia

Jul 2022 – Oct 2023

10TB+ daily for 1.7M customers

  • Built Databricks Delta Lake pipelines on a bronze/silver/gold medallion architecture in PySpark and Spark SQL.
  • Tuned performance with adaptive query execution and partition strategy, and replaced a legacy Oracle Data Integrator estate with a Python ETL framework.
  • Migrated an on premises Oracle warehouse to AWS Redshift.

Senior Consultant, Data Designer & EngineerNational Australia Bank

Jun 2019 – Jul 2022

Big Four bank, billions in daily transactions

  • Designed live ingestion of call centre events, and an AWS data lake on S3, Glue and Athena for unstructured analysis.
  • Implemented change data capture for continuous warehouse updates, and dimensional models supporting 50+ executive KPIs.
  • Established data governance practices for Open Banking compliance.

Enterprise data and ETLRepublic Services · Discount Tire · Fortune 500 clients, US

2005 – 2019
  • Built an enterprise operational data store integrating 25+ source systems, led a Denodo data virtualisation rollout, and developed 500+ Informatica mappings with continuous CDC via PowerExchange.
  • Delivered master data management, a SAP migration across 900+ retail sites, and an AS/400 to SQL Server modernisation.
  • Led data engineering teams of up to 30 across onshore and offshore.

Skills

AI
Azure OpenAI, RAG, evaluation and guardrails, LangChain, LlamaIndex, vector databases (Pinecone, Chroma, FAISS), prompt engineering, Hugging Face
Data
Databricks, Delta Lake, PySpark, Spark SQL, Kafka, Airflow, Informatica, change data capture, dimensional modelling
Cloud
Azure (Data Factory, Synapse, ADLS, Functions), AWS (Redshift, Glue, S3, EMR)
Languages & stores
Python, SQL, PL/SQL, FastAPI, SQL Server, Oracle, PostgreSQL, Neo4j

Education & certifications

2018
MS Computer Science, machine learning specialisation. Georgia Institute of Technology.
2005
BTech Information Technology. Mahatma Gandhi University, India.
Certified
Databricks Certified Data Engineer Associate (2024).