Data Scientist - AI [T500-27599]

Talent500; India; On-site; INR 900,000 - 1,300,000; Full time

Job summary

Talent500 seeks a Data & AI Analyst who can bridge traditional analytics with Generative AI. You will parse highly unstructured documents, orchestrate AI models, and drive automation across large datasets using cloud storage formats like Parquet and vector databases.

You will select tools, build RAG architectures, integrate AI outputs into business workflows, and write clean code with Git. Strong Python, SQL, and cloud OCR experience are essential.

Qualifications

• Hands-on experience parsing messy or unstructured files and extracting data.
• Familiarity with Azure Document Intelligence or equivalent extraction tools.
• Practical experience building Retrieval-Augmented Generation systems with vector databases.
• Familiarity with commercial APIs (OpenAI, Anthropic, Gemini) and open-source models.

Responsibilities

• Parse unstructured documents to extract structured insights.
• Evaluate tech options and select data libraries, OCR tools, or LLMs.
• Integrate AI models with business apps and workflows.
• Automate data-driven reports and executive briefs.
• Query and transform large datasets; optimize storage formats like Parquet.
• Design and maintain RAG architectures and Azure Document Intelligence usage.
• Maintain clean, documented code and collaborate via Git workflows.

Skills

Unstructured document parsing

RAG architectures

Python

SQL

APIs & integration

Git version control

LangChain/LlamaIndex

Tools

Azure Document Intelligence

LangChain

LlamaIndex

OpenAI API

Anthropic API

Gemini API

Snowflake

Parquet

Job description

Role Overview

We are seeking a versatile and forward‑thinking Data & AI Analyst to join our team. This role bridges the gap between traditional data analytics and modern Generative AI implementation. The analyst will parse highly unstructured documents, leveraging advanced language models to drive automation and insights while handling large‑scale datasets. Strong engineering fundamentals and problem‑solving skills are valued over platform‑specific certifications.

Key Responsibilities

• Unstructured Document Parsing: Extract clean, structured insights from a wide variety of complex files, including PDFs, multi‑tab Excel workbooks, financial models, images, and scanned forms.
• Tool Selection & Strategy: Evaluate business problems and select the optimal technology mix, choosing standard data libraries, cloud OCR tools, or large language models.
• Business Application Integration: Bridge standalone AI models with core business applications, ensuring extracted data and insights flow smoothly into internal workflows and systems.
• Automated Report Writing: Design and implement systems that automate the creation of data‑driven reports, summaries, and executive briefs, combining structured data analytics with Generative AI text generation.
• Data Management & Processing: Query, transform, and manage large datasets within cloud data warehouses, using optimized data storage formats such as Parquet for high‑performance pipelines.
• AI & RAG Architecture: Design, build, and maintain Retrieval‑Augmented Generation (RAG) systems to ground LLMs in internal company data, utilizing tools like Azure Document Intelligence for upstream extraction.
• Collaboration & Code Hygiene: Maintain clean, documented, and reusable codebases; participate in team workflows using standard Git version control practices.

Required Skills & Qualifications

Artificial Intelligence & Document Parsing

• Advanced Document Extraction: Hands‑on experience extracting data from messy, semi‑structured, or completely unstructured files, including programmatic parsing or chunking of complex Excel/spreadsheet data and PDFs.
• Extraction Tools: Familiarity with Azure Document Intelligence or equivalent advanced layout extraction tools.
• RAG Frameworks: Practical experience building Retrieval‑Augmented Generation systems, working with vector databases and frameworks such as LangChain or LlamaIndex.
• Model Selection: Familiarity with commercial APIs (OpenAI, Anthropic, Gemini) and lightweight/open‑source models (Llama, Mistral, Phi).

Integration & Software Engineering

• Systems Integration: Experience building and consuming RESTful APIs to connect AI models, data systems, and downstream business applications.
• Version Control: Proficiency with Git and standard platform workflows (GitHub, GitLab, or Azure DevOps).

Data & Analytics Infrastructure

• Big Data Handling: Comfortable working with large‑scale datasets and optimized data formats (JSON, CSV, Parquet).
• Data Pipelines: Strong proficiency manipulating data via Python and writing clean, efficient SQL.

Preferred / Nice‑to‑Have Qualifications

• Experience within the Azure cloud ecosystem and running queries in Snowflake.
• Experience building systems that automatically synthesize data into natural language reports or automated client‑facing summaries.
• Familiarity with prompt engineering techniques and LLM evaluation frameworks.