Abhishek Das

Hey, I'm Abhishek

I build data pipelines, tinker with GenAI agents, and write about gaming & tech on this blog. Currently based in Hyderabad, turning messy data into things that actually work.

I've been working with data for over 4.5 years now. I started out wrangling Spark jobs and Azure pipelines, figuring out how to move data from point A to point B without everything breaking at 2 AM. Over time I fell in love with the craft of building things — fast APIs, reliable pipelines, systems that quietly do their job.

More recently, I dove headfirst into the GenAI world — building agentic applications with LangChain and LangGraph, working with vector indexes, and adding observability so you can actually see what your LLM is doing instead of hoping for the best. It's the most fun I've had in this field.

Tiger Analytics Jan 2025 — Present

GenAI Developer

This is where things got really interesting. I'm building and shipping GenAI-powered applications — think LLM-based agents that can reason, search, and act. I work with LangChain and LangGraph to orchestrate agentic workflows, set up vector indexes for semantic search and retrieval, and built out the FastAPI layer that serves it all. One thing I'm especially proud of: adding observability to our LLM and agentic pipelines — tracing calls, tracking token usage, monitoring latency — so the team can debug and iterate fast instead of flying blind.

LangChain LangGraph Vector Indexes FastAPI LLM Observability Python
Deloitte Jul 2023 — Jan 2025

Consultant Data Engineer

Built a Data Exchange framework in PySpark that cut processing time by 15% and auto-created ServiceNow tickets when things went sideways. Wired up Azure Databricks jobs to trigger from Data Factory with multi-threaded parallelism — shaved 20% off computation time. Also built a Query Generator that turned XML configs into SQL, which meant fewer manual errors and way faster data retrieval.

PySpark Azure Databricks Azure Data Factory Delta Lake ServiceNow API
Cognizant Jul 2021 — Jul 2023

Data Engineer

Where it all started. Learned to think in DataFrames, built event-triggered ADF pipelines that cut ingestion latency by 25%, and wrote the bronze/silver layer scripts that became the backbone of our data lake. Two years of getting really comfortable with Spark, S3, and the feeling of a pipeline finally running green.

PySpark AWS S3 Azure Data Factory SQL
Python PySpark FastAPI LangChain LangGraph Vector Databases LLM Observability Azure (ADF, ADLS, Databricks) AWS (S3, Redshift) Delta Lake PostgreSQL MongoDB Django SQL Docker
Building

GenAI agents & this blog you're reading

Exploring

Agentic RAG, multi-agent orchestration

Writing

Gaming reviews, tech deep-dives, tutorials

Education

B.Tech CS — C.V. Raman College (8.7 CGPA)

Email copied to clipboard!