Pranaav Srinivasan

Stack

The tools I use, in the order data meets them: from the first event to the dashboard that shows whether it all worked.

  1. Ingest

    Getting events in

    • Apache Kafka
    • Kinesis
    • SQS
    • REST APIs

  2. Process

    Batch and streaming

    • Apache Spark
    • PySpark Structured Streaming
    • Hadoop
    • Dask
    • Ray
    • Pandas
    • NumPy
    • PyArrow

  3. Orchestrate

    Scheduling and recovery

    • Apache Airflow
    • GitHub Actions
    • Jenkins

  4. Store and model

    Lakes, warehouses, vectors

    • AWS S3
    • Snowflake
    • PostgreSQL
    • MongoDB
    • Redis
    • Apache Hive
    • ChromaDB
    • Pinecone
    • Neo4j
    • Medallion architecture

  5. Serve

    APIs and interfaces

    • FastAPI
    • Flask
    • REST
    • Spring Boot
    • Nginx
    • Flutter
    • Power BI

  6. Observe

    Knowing it works

    • Amazon CloudWatch
    • Grafana
    • Prometheus
    • ELK Stack
    • MLflow

Across the stack

Languages

  • Python
  • SQL
  • Java
  • Scala
  • Rust
  • C
  • C++
  • JavaScript
  • HTML

ML and AI

  • PyTorch
  • TensorFlow
  • Hugging Face
  • scikit-learn
  • LangChain
  • LangGraph
  • RAG
  • Pydantic

Cloud and DevOps

  • AWS (S3, EC2, IAM)
  • Docker
  • Terraform
  • Ansible
  • Linux
  • Git