Available for opportunities

Anushka
Shukla
AI & Data Engineer

Building intelligent systems and scalable AI pipelines. Crafting multi-agent platforms, data validation frameworks, and knowledge-graph powered applications at enterprise scale.

↗ Resume Get in Touch
About Me

Turning ideas into
intelligent systems

A fast-learning AI & Data Engineer with hands-on experience building enterprise-grade multi-agent platforms, data pipelines, and knowledge-graph applications.

I'm Anushka — an AI & Data Engineer currently at Accellor Inc., working on enterprise multi-agent AI automation platforms and scalable data validation systems. My work spans the full stack from architecting orchestration graphs with 15+ specialized agents to building robust Human-In-The-Loop pipelines and deploying cloud-native solutions on Azure.

With a B.Tech in CSE (AI/ML specialization) from VIT Bhopal, I bring a strong academic foundation in deep learning and machine learning, paired with real-world experience in LLM integration, knowledge graphs, and enterprise data engineering.

15+
AI Agents Built
60%
Automation Gain
4
Certifications
3+
Enterprise Projects
Anushka Shukla
🏢
Accellor Inc.Data & AI Engineer
Experience

Where I've worked

Enterprise-scale AI systems, data pipelines, and multi-agent platforms.

Data & AI Engineer
Jul 2024 – Present
Accellor Inc. · Hyderabad, Telangana
G42 AI Agents — Enterprise Multi-Agent Automation Platform
  • Architected and deployed a multi-agent automation platform using FastAPI, Microsoft Agent Framework (MAF), and OpenAI-powered agents for end-to-end enterprise content generation including press releases, social media, and localized communications.
  • Built a scalable orchestration graph with dynamic agent routing (Orchestrator → Dispatcher → Redirector) coordinating 15+ specialized agents, cutting manual content creation time by ~60%.
  • Implemented a robust Human-In-The-Loop (HITL) review pipeline with reviewer agents and feedback handlers, reducing manual editing effort by an estimated 50–60%.
  • Developed modular agent executors with feedback-driven revision cycles via structured agent-reviewer loops (PressRelease ↔ BrandReviewer, TwitterAgent ↔ TwitterReviewer, Localizer ↔ LocalizerReviewer).
  • Built unified API endpoints for 10+ workflows with automated DB-level logging for all agent events, HITL actions, and outputs — improving traceability and auditability.
FastAPIOpenAI APIMicrosoft Agent Framework Multi-Agent SystemsHITL PipelinesPython
Data Validation Platform — Claire's Inc.
  • Led data migration and validation for Claire's Inc., overseeing transfers from PMM Lakehouse, Microsoft Fabric, AS400, XStore, Salesforce OMS API, and Azure Blob Storage — ensuring seamless integration and data integrity across diverse systems.
  • Built and deployed an Azure Function App to automate and orchestrate the entire validation pipeline, significantly reducing manual effort.
  • Designed a scalable data validation platform using Great Expectations, enabling automated accuracy, completeness, and consistency checks across SQL, CSV, XML, and API sources.
Azure Function AppsGreat ExpectationsMicrosoft Fabric SalesforceData MigrationPython
AstraZeneca — Global Project Portfolio Management
  • Architected a FastAPI-based multi-agent AI system for AstraZeneca's GPPM division, enabling intelligent document processing, semantic search, and real-time chat over R&D and financial data.
  • Integrated multiple specialized ReAct agents (Project, Finance, Milestone, Summariser) using LangGraph and Azure OpenAI, orchestrated by a supervisor agent for dynamic query routing and consolidation.
  • Implemented Neo4j knowledge graph integration with vector embeddings for semantic relationships across project, financial, and milestone data — powering contextual search and reasoning.
  • Connected the system with PostgreSQL for structured retrieval and Opik tracing for performance monitoring and debugging.
LangGraphAzure OpenAINeo4j ReAct AgentsVector EmbeddingsPostgreSQLFastAPI
Skills

What I work with

A broad toolkit spanning AI/ML, cloud platforms, data engineering, and full-stack development.

🧠
AI & Machine Learning
  • LLM Integration & Agents
  • Deep Learning & ANNs
  • Computer Vision
  • TensorFlow
  • LangChain / LangGraph
☁️
Cloud & Infrastructure
  • Azure Function Apps
  • Azure Blob & Event Hubs
  • Microsoft Fabric
  • Google Cloud Platform
  • Docker / Deployment
🗄️
Data Engineering
  • Great Expectations
  • Neo4j Knowledge Graphs
  • PostgreSQL / SQL
  • Vector Embeddings
  • Salesforce / APIs
💻
Languages & Tools
  • Python
  • JavaScript / TypeScript
  • C++
  • FastAPI / REST APIs
  • Git / GitHub
Projects

Things I've built

A selection of personal and collaborative projects showcasing applied AI and software engineering.

🤚 01
Sign Language to Text

A real-time fingerspelling recognition system using a custom CNN trained on a self-captured dataset of all 26 ASL alphabet gestures. Employs a two-layer classification pipeline — a primary CNN classifier combined with secondary specialized classifiers for visually similar gestures — achieving 98% accuracy. Ships with a full Tkinter GUI for live webcam-based gesture-to-sentence conversion.

CNNTensorFlowOpenCVKerasTkinterComputer Vision
🎵 02
Spotify Music Analysis

EDA and feature engineering on Spotify song metadata — uncovering patterns in audio features, genre clusters, and popularity trends using statistical analysis and visualizations.

PythonJupyterPandasEDA
↗
📷 04
Smart Lens — OCR App

An Android application leveraging AI-powered Optical Character Recognition to detect, extract, and digitize text from photos in real time — built with a Python ML backend.

Android StudioPythonOCRComputer Vision
↗
Certifications

Credentials

DB
Data Engineer Associate
Databricks
ML
Machine Learning Specialization
Stanford University / Coursera
AI
Azure AI Fundamentals (AI-900)
Microsoft
GC
Big Data & ML Fundamentals
Google Cloud
Contact

Let's work together

I'm always open to new opportunities, collaborations, or just a good conversation. Drop me a line.

📧 Email Me LinkedIn
© 2025 Anushka Shukla · All rights reserved