Skip to content
View Mohith-akash's full-sized avatar

Block or report Mohith-akash

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Mohith-akash/README.md

Mohith Akash

Data Engineer & GenAI Developer

LinkedIn · Email · Germany

Building production data platforms and LLM-powered systems. 3 live projects, 30M+ records processed, 8,600+ users.

About me

I ran MAS Nexus, my own engineering practice, for over two years: 27 months as an embedded data engineer for a client on vendor basis (data pipelines, reporting layer, tooling decisions, AI features), plus the platforms below built alongside. Reference letter available on request.

Based in Germany as a Chancenkarte (Opportunity Card) holder, authorized to work anywhere in Germany and available immediately.

Projects

Serverless ELT pipeline over 30M+ geopolitical news events from GDELT, live since December 2025.

  • 8,600+ unique visitors since launch, zero advertising
  • 100K+ events ingested daily on an hourly refresh cycle, dbt marts with 30+ tests built weekly
  • Two-mode AI chat: RAG over vector embeddings, plus SQL answers from tested query templates
  • Slowest stage rewritten in Polars (~10x faster than Pandas)
  • $0/month: MotherDuck + Dagster + GitHub Actions on free tiers

Python Polars dbt MotherDuck Dagster Great Expectations

Real-time streaming platform for AI-powered cart-abandonment recovery.

  • Azure Event Hubs streaming into Delta Live Tables (Bronze, Silver, Gold)
  • LLM-personalized recovery messages across 5 customer archetypes
  • Medallion lakehouse (Bronze/Silver/Gold) with DLT expectations
  • A/B testing with z-scores at 95% confidence

Databricks Azure Event Hubs DLT Cerebras dbt Streamlit

Databricks lakehouse over 100K+ orders from the Brazilian Olist marketplace.

  • Medallion architecture with Delta Lake
  • Kimball star schema: fct_orders + customer/product/seller dimensions
  • Unity Catalog governance with documented lineage

Databricks Delta Lake SQL Unity Catalog Power BI

Anki deck with 5,000+ German words across A1 to C1, built while learning German myself. 7,800+ downloads on AnkiWeb, all organic.

  • Designed the build pipeline with AI: word selection by frequency and learner need
  • Neural audio for every word, images, verb conjugations

Python Anki Text-to-Speech

Interactive Excel dashboard analyzing 12,894 data job postings.

  • 70+ step Power Query ETL, 228K+ rows in Power Pivot with custom DAX
  • 4,000+ messy job titles standardized into role categories

Excel Power Query Power Pivot DAX

Tools

  • Data engineering: Python · SQL · Databricks · Delta Lake · Azure · DuckDB/MotherDuck
  • Pipelines: dbt Core · Dagster · Polars · Great Expectations · GitHub Actions
  • AI/LLM: RAG · LlamaIndex · Cerebras · Voyage AI embeddings
  • Visualization: Streamlit · Plotly · Power BI

Certifications


Data Engineer & GenAI Developer. Open to Data Engineer, Analytics Engineer, AI Engineer, and Data Analyst roles in Germany. Available immediately, authorized to work in Germany (Chancenkarte, no sponsorship needed), English C1, German A2 (working toward B1).

B.Tech, Electrical & Electronics Engineering, SRM University, Chennai

Pinned Loading

  1. Global-News-Intel-Platform Global-News-Intel-Platform Public

    Geopolitical news platform on GDELT: 30M+ events, about 100K added daily. Polars, Dagster and Great Expectations load MotherDuck hourly, dbt builds tested marts weekly; AI chat answers with RAG and…

    Python 27 4

  2. Vortex-The-Revenue-Recovery-Engine Vortex-The-Revenue-Recovery-Engine Public

    Real-time cart-abandonment recovery on Azure Event Hubs and Databricks: Delta Live Tables under 500ms end to end, and an LLM writes a recovery message for each abandoned cart.

    Python

  3. olist-analytics-platform olist-analytics-platform Public

    End-to-end analytics platform: CSV → Databricks → Delta Lake → Streamlit Dashboard | 100K+ Brazilian e-commerce orders

    Python

  4. german-frequency-deck german-frequency-deck Public

    5,009-word German Anki deck (A1 to C1) with neural audio, images, full verb conjugations, Goethe level badges and word families

    Python 12