---
title: "Solon Das - Data & AI Engineering - Palantir Foundry in Bengaluru | Tal"
description: "Meet Solon Das, Data & AI Engineering @Swiss Re . Palantir Foundry . AI Infrastructure & Data Platforms . Spark . DataBricks . Azure . AWS based in Bengaluru. AI stack includes antigravity, claude code, and claude. Explore public work links including github, medium, and portfolio-dznk.vercel.app."
image: "https://tal.club/meet/solon-kumar-das/share-image"
---

# Solon Das on Tal

## Headline

Data & AI Engineering @Swiss Re \. Palantir Foundry \. AI Infrastructure & Data Platforms \. Spark \. DataBricks \. Azure \. AWS

## About

I build the data and AI infrastructure that powers intelligent enterprises\. With 7\+ years of experience across FMCG, Financial Services, Healthcare, and Reinsurance, I specialize in designing large\-scale data platforms, AI\-ready architectures, and knowledge systems that transform enterprise data into business value\. My work sits at the intersection of Data Engineering, Platform Engineering, and Generative AI\. I have built and scaled TB\-scale data ecosystems, real\-time streaming platforms, enterprise data products, and multimodal AI solutions that enable analytics, automation, and AI\-driven decision making\. Key achievements: \- Architected and delivered production data platforms powering enterprise GenAI products across multiple global business regions\. \- Led the consolidation of fragmented document ecosystems into a governed platform spanning 85M\+ documents and 59TB\+ of enterprise data, enabling large\-scale analytics, search, and AI use cases\. \- Built multimodal AI capabilities using Vision\-Language Models \(VLMs\), vector embeddings, semantic search, and Retrieval\-Augmented Generation \(RAG\) to unlock insights from unstructured enterprise content\. \- Developed data products that contributed to a 5% incremental sales uplift and $1\.7M net revenue growth across global Direct\-to\-Consumer markets\. \- Reduced monthly platform costs by 51\.2%, from $26\.9K to $13\.1K, by profiling Spark execution plans, replacing expensive UDFs with native Spark functions, and implementing incremental processing strategies\. \- Designed real\-time, event\-driven architectures using Spark, Kafka, Azure Databricks, Azure Data Factory, Azure Cosmos DB, and Palantir Foundry to support mission\-critical workloads at scale\. My core areas of expertise include: • AI Infrastructure & GenAI Data Platforms • RAG, Semantic Search & Knowledge Systems • Vector Databases & Enterprise Search • Multimodal AI & Vision\-Language Models • Apache Spark & Distributed Data Processing • Azure Databricks & Palantir Foundry • Real\-Time Streaming Architectures • Data Platform Modernization • Data Governance, Quality & Observability • Platform Performance & Cost Optimization I am passionate about building systems that make enterprise data discoverable, trustworthy, and AI\-ready\. Whether it's architecting data platforms, enabling natural language access to business knowledge, or optimizing large\-scale distributed workloads, I enjoy solving complex engineering challenges that create measurable business impact\. Let's chat \- https://portfolio\-dznk\.vercel\.app

## Location

Bengaluru

## AI stack

antigravity

claude code

claude

gemini

github copilot

openai

## Work experience

Data & AI Engineering \- Palantir Foundry at Swiss Re, Mar 2025 \- Present · 1 yr 7 mos, Bengaluru, Karnataka, India, Built and Scaled "AnalytiC Document Corpus \(ACDC\)" \- Palantir Foundry 1\. Led and Designed cross\-functional delivery across core SwissRe business CorSo, L&H, and P&C to consolidate fragmented document ecosystems into a single governed platform \(\~90M documents / \~59TB\), establishing a consistent, compliant access model and reducing reliance on siloed, team\-specific pipelines 2\. Defined and drove adoption of standardized document processing \(PDF, OCR, text\) and centralized enrichment pipelines, enabling reuse of extracted insights across use cases, reducing duplicated compute and engineering effort, and accelerating ML/GenAI experimentation and deployment 3\. Operationalized request based ingestion from core systems, ensuring up\-to\-date data availability for analytics and decision\-making 4\. Optimized ACDC Spark workloads to reduce monthly platform cost by 51\.2%, from $26\.9K in January to $13\.1K by April, with the largest month\-on\-month reduction of 42\.8% achieved in April by profiling Spark execution plans, replacing costly UDFs with Spark native functions and shifting to incremental loads "Unstructured Data Model \(UDM\)", Asset Management \- Palantir Foundry 1\. Led the architecture and delivery of AM\-UDM on Palantir Foundry, a multimodal data product powered by Vision\-Language Models \(VLMs\), forming the foundational AI layer for Generative AI use cases across Asset Management 2\. Built a scalable multimodal extraction pipeline to parse text, images, charts, and tables from heterogeneous sources \(e\.g\., Outlook emails, user uploads\), handling noisy unstructured data at enterprise scale 3\. Developed a contextual summarization layer using vision models to generate semantically rich summaries from combined visual and textual content, significantly improving downstream LLM performance 4\. Built semantic search using vector embeddings, enabling deep retrieval across the document corpus and powering a chat\-based interface for analysts to uncover insights from unstructured data

Tech Blogger \- TheDatumTheory at Medium, Jan 2024 \- Present · 2 yrs 9 mos, Remote, Most data systems don’t fail because of theory, they fail because of poor decisions in the real world\. That’s exactly what I write about on my Blogs\. I focus on data engineering beyond the textbook: • What actually breaks in production pipelines • Why Spark jobs slow down and how to fix them • The trade\-offs behind scalable data architectures • Designing systems that don’t just work, but keep working This isn’t about repeating documentation, it’s about challenging assumptions, sharing hard\-earned lessons, and building systems that scale under pressure\. If you're serious about data engineering and want perspectives grounded in reality, take a look:

Senior Data Engineer \- Generative AI at AB InBev, Sep 2022 \- Mar 2025 · 2 yrs 7 mos, Bengaluru, Karnataka, India, Senior Data Engineer – Generative AI & Data Platforms Project: Insights Copilot 1\. Architected and delivered end\-to\-end data pipelines for Insights Copilot, an enterprise GenAI platform enabling interactive, cross\-functional querying across business operations, serving as the primary data engineering lead from inception to production launch\. 2\. Designed and operationalized a multi\-domain Data Quality Framework spanning FMCG business domains, reducing downstream data issues by proactively detecting and resolving ingestion anomalies before reaching GenAI application layers\. 3\. Drove NL2SQL solution design and data strategy across 8\+ global business functions \(Marketing, Finance, HR, Logistics, Procurement, Social Media, Strategy & Insights, Watchtower\) and 5 regions \(Global HQ, APAC, NAZ, SAZ, MAZ\), enabling scalable natural language querying for diverse enterprise stakeholders\. 4\. Engineered reusable data models and enriched metadata schemas to expand LLM context windows, directly optimizing inference quality and consistency across 5\+ multi\-domain GenAI use cases\. Data Engineer – DTC Data Delivery Pipelines South America Markets 1\. Built production\-grade, end\-to\-end data pipelines on Azure Databricks from the ground up to power the DTC\-MROI \(Direct\-to\-Consumer Marketing Return on Investment\) segment across Brazil, Mexico, Colombia, and Argentina, enabling data\-driven marketing decisions in high\-growth emerging markets\. 2\. Delivered market\-localized data solutions within a $50M budget, contributing to a projected 5% incremental sales uplift and $1\.7M net revenue gain through trusted, high\-value data delivery to downstream analytics and business teams\.

Big Data Engineer \- Senior Consultant at EY, May 2021 \- Sep 2022 · 1 yr 5 mos, Bengaluru, Karnataka, India, Real\-Time Streaming & Event\-Driven Data Infrastructure 1\. Architected and delivered an end\-to\-end Event\-Driven Ingestion Pipeline on Azure Databricks using Apache Spark, orchestrating real\-time data flow across Azure Cosmos DB and Apache Kafka enabling low\-latency, scalable ingestion for high\-throughput enterprise workloads\. 2\. Engineered a Data Tracing & Observability Feature that tracked pipeline load statuses end\-to\-end, persisting audit records to SQL Server and publishing events to Azure Event Hub providing full operational visibility into ingestion health and reducing mean time to detect pipeline failures\. 3\. Built a real\-time Kafka Message Recovery System that enabled rapid identification and replay of lost or failed messages across topics, significantly improving pipeline resilience and ensuring data completeness in event\-driven workflows\.

Data Engineer at Accenture, Jul 2018 \- Nov 2020 · 2 yrs 5 mos, Pune District, Maharashtra, India, Data Engineer \- Tooling, Automation & Data Transformation 1\. Engineered an Object Management Tool for Environment Deployment, standardizing release workflows across environments and eliminating manual handoffs for a team supporting large\-scale pharmaceutical data infrastructure\. 2\. Built an Automatic JIL File Generator in Python that automated scheduling configuration for 30\+ production pipelines, reducing manual job scheduling effort by 98% and significantly lowering the risk of misconfiguration in a regulated pharmaceutical data environment\. 3\. Led a PySpark Data Transformation initiative for Merck’s Rx \(prescription drug\) data platform, processing terabyte\-scale datasets to integrate new prescription drug data points into client\-specific transformation pipelines, enabling accurate, compliant downstream reporting across the non\-prescription and Rx drug categories

## Education

Master of Science \- MS, Data Science, Liverpool John Moores University

Post Graduate Diploma, Data Science, International Institute of Information Technology Bangalore

## Public links

- github: https://github\.com/solondas96
- medium: https://medium\.com/@solondas96
- portfolio\-dznk\.vercel\.app: https://portfolio\-dznk\.vercel\.app/

[View the original profile](https://tal.club/meet/solon-kumar-das)

```json
{
  "@context": "https://schema.org",
  "@graph": [
    {
      "@type": "Person",
      "@id": "https://tal.club/meet/solon-kumar-das#person",
      "name": "Solon Das",
      "url": "https://tal.club/meet/solon-kumar-das",
      "image": "https://tal.club/meet/solon-kumar-das/profile-photo.jpg",
      "description": "Meet Solon Das, Data & AI Engineering @Swiss Re . Palantir Foundry . AI Infrastructure & Data Platforms . Spark . DataBricks . Azure . AWS based in Bengaluru. AI stack includes antigravity, claude code, and claude. Explore public work links including github, medium, and portfolio-dznk.vercel.app.",
      "jobTitle": "Data & AI Engineering - Palantir Foundry",
      "worksFor": {
        "@type": "Organization",
        "name": "Swiss Re"
      },
      "hasOccupation": {
        "@type": "Occupation",
        "name": "Data & AI Engineering - Palantir Foundry"
      },
      "homeLocation": {
        "@type": "Place",
        "name": "Bengaluru"
      },
      "alumniOf": [
        {
          "@type": "EducationalOrganization",
          "name": "Liverpool John Moores University"
        },
        {
          "@type": "EducationalOrganization",
          "name": "International Institute of Information Technology Bangalore"
        }
      ],
      "knowsAbout": [
        "antigravity",
        "claude code",
        "claude",
        "gemini",
        "github copilot",
        "openai"
      ],
      "sameAs": [
        "https://github.com/solondas96",
        "https://medium.com/@solondas96",
        "https://portfolio-dznk.vercel.app/"
      ],
      "mainEntityOfPage": {
        "@id": "https://tal.club/meet/solon-kumar-das#webpage"
      }
    },
    {
      "@type": "ProfilePage",
      "@id": "https://tal.club/meet/solon-kumar-das#webpage",
      "url": "https://tal.club/meet/solon-kumar-das",
      "name": "Solon Das on Tal",
      "description": "Meet Solon Das, Data & AI Engineering @Swiss Re . Palantir Foundry . AI Infrastructure & Data Platforms . Spark . DataBricks . Azure . AWS based in Bengaluru. AI stack includes antigravity, claude code, and claude. Explore public work links including github, medium, and portfolio-dznk.vercel.app.",
      "dateModified": "2026-09-08T16:41:05.902Z",
      "isPartOf": {
        "@id": "https://tal.club/#website"
      },
      "mainEntity": {
        "@id": "https://tal.club/meet/solon-kumar-das#person"
      },
      "primaryImageOfPage": {
        "@type": "ImageObject",
        "url": "https://tal.club/meet/solon-kumar-das/profile-photo.jpg",
        "contentUrl": "https://tal.club/meet/solon-kumar-das/profile-photo.jpg",
        "caption": "Solon Das profile photo on Tal"
      },
      "inLanguage": "en-IN"
    }
  ]
}
```