todo-bytes
A local-first task manager backed by plain YAML, accessible through a CLI, browser UI, AI-agent skill, and MCP server.
View source
profile/context.md
context loaded
Hi, I'm Rishi Tamrakar
Data Engineering · Cloud · Agentic AI
I build reliable data platforms and practical AI agents. This profile traces 10+ years from legacy ETL to cloud-native systems and evidence-driven automation.
10+Years in Data
GCP & AWSCloud Native
🇩🇪Based in Germany
// agent profile
Rishi Tamrakar
Working across data, cloud, and agents
data/Data Engineeringpipelines
cloud/GCP · AWSplatforms
agents/Agentic AIactive
$ build --with-context
About
From legacy ETL to intelligent data platforms
I've lived through the full evolution of data engineering, from building Informatica PowerCenter mappings to designing modern, cloud-native data platforms on GCP and AWS. I've seen what works at scale and what quietly breaks at 3 AM.
Today, I work at the intersection of data engineering and Agentic AI, exploring how agents can support investigation, automation, and faster problem resolution. I design scalable architectures (streaming & batch), migrate legacy systems to modern stacks, and build data platforms that teams can actually trust.
I also write and create content about the real challenges in data engineering, architecture trade-offs, career pivots, the honest side of working with data at scale, and life in Germany as an expat.
Journey
A decade across data engineering
From legacy ETL to cloud-native platforms and AI-assisted engineering.
-
Foundation TCS
Legacy ETL foundations
Building enterprise data workflows with Informatica PowerCenter.
-
Scale Clairvoyant / PayPal
Big data at scale
Moving into distributed data processing with Hadoop and Spark.
-
Cloud Uplight
Cloud data platforms
Building reliable, maintainable systems on modern cloud stacks.
-
Intelligence FINN
Data meets Agentic AI
Evolving GCP data platforms with practical AI-assisted workflows.
Now — building reliable data platforms and exploring practical Agentic AI.
What I talk about
Topics & themes
Data Engineering
Scalable pipelines, dbt, Spark, Kafka, Airflow. Streaming & batch architectures that work at scale.
Cloud Platforms
Building on GCP & AWS. BigQuery, Dataflow, Pub/Sub, and designing cloud-native data platforms.
Agentic AI
Exploring how AI agents support investigation, automation, and problem resolution in modern data platforms.
Career & Growth
Career strategies, job switches, and practical advice for data engineers at any level.
Life in Germany
What it's really like. Visas, work culture, probation, and settling in as an expat.
Toolbox
Tech I work with
The tools and platforms behind the systems I build.
Languages & Processing
Python
SQL
Spark
dbt
Kafka
Airflow
Cloud & Data Platforms
GCP
AWS
BigQuery
Dataflow
Pub/Sub
Snowflake
AI & Automation
Agentic AI
LLMs
RAG
Vector DBs
Terraform
CI/CD
Selected builds
Things I build and explore
Open-source projects and practical experiments across data and AI.
OpenClaw Docker Lab
A sandboxed Docker starter for running and personalising an AI assistant locally, with setup and security guidance.
View source
Videos & talks
Ideas in conversation
Workshops and conversations I've had the opportunity to contribute to, alongside the latest videos from Rishi Bytes.
Community workshop · MSMV Group
AI Workshop — History, Present & Future
I had the opportunity to contribute to an MSMV Group session exploring the history, present, and future of artificial intelligence.
Latest from Rishi Bytes
Loading videos...
Writing
From the blog
Practical field notes on data engineering, architecture, Agentic AI, and engineering careers.
Cheap Tokens Won’t Make You a Better Engineer
AI usage may become cheaper or more expensive, but engineering value still comes from solving the right problem within real constraints.
July 10, 2026 · 4 min read
Why I’m Breaking the DRY Rule: The Case for Data Duplication
Data duplication is not automatically technical debt. Sometimes separate copies are the clearest way to meet different latency, scale, and analytical requirements.
March 30, 2026 · 3 min read
I Stopped Googling. Is AI Search Actually Honest?
AI answers feel cleaner than search results, but recommendations can still be shaped by training data, commercial integrations, and advertising.
March 15, 2026 · 4 min read
Let's connect
Building something with data or AI?
Whether it's a data platform, a cloud migration, or a practical AI-assisted workflow, I'm always happy to chat. Find me on any of these.