Skip to content
View akankshakusf's full-sized avatar

Block or report akankshakusf

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.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

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

Report abuse
akankshakusf/README.md

💫 About Me:

Hi, thanks for stopping by! I’m a data-driven builder with 6+ years of experience in data engineering, now immersed in the world of Gen AI, Deep Learning and Machine Learning. I specialize in crafting intelligent systems that combine structured enterprise data with modern AI to deliver scalable, production-ready solutions.

I design MCP Autonomous Agents with OpenAI framework, Agentic RAG pipelines using LangChain framework, use FAISS, ChromaDB, AstraDB Vector Store DBs and API-based augmentation to power real-time, context-aware search and Q&A tools. I monitor and debug these pipelines with LangSmith to ensure consistent and reliable performance.

I build Multi-Agent frameworks with CrewAI, LangGraph, AutoGen & Strands— orchestrating tool-using agents that reason, collaborate, and delegate tasks. With CrewAI, I’ve built role-specific agents for research, summarization, and tool execution, enabling intelligent workflows across company use cases.

My multi-model systems are integrated with AWS Bedrock, LangChain, and dynamic routing logic to switch seamlessly between Open AI, Claude, Grok etc. I use session-aware orchestration, S3 for storing artifacts, and deploy models via Elastic Beanstalk and containerized services.

In Machine Learning, I’ve worked on models for churn prediction, dynamic pricing, sentiment classification, and fraud detection using Random Forest, XGBoost, Decision Trees, Logistic Regression, & Gradient Boosting. I use ML Ops Weights & Biases (WandB), Langfuse, Tensorboard for deep experiment tracking, metric visualization, artifact management, and collaborative model comparison across iterations.

On the Deep Learning side, I perform image generation with Stable Diffusion, neural machine translation with T5 and MarianMT, and built speech recognition systems using Whisper. I’m also hands-on with RNNs for sequence tasks, and CNNs for classification. I frequently prototype in Google Colab & deploy using SageMaker for scalable training, model hosting, and pipeline automation.

I’m passionate about building solutions that are smart, scalable, & ready for enterprise environments.

Tools & Technologies: LangChain, LangGraph, CrewAI, Lamini, FAISS, ChromaDB, AstraDB, Neo4j, OpenAI, Claude, Groq, Gemma, AWS Bedrock, Amazon S3, Elastic Beanstalk, SageMaker, Google Colab, Power BI, Stable Diffusion, Whisper, T5, MarianMT, Random Forest, XGBoost, Decision Trees, Logistic Regression, Gradient Boosting, CNNs, RNNs, MLflow, Weights & Biases (WandB), LangSmith, Python, Azure Data Factory, SQL, Power Automate, DAX, SSIS, SSRS, SSAS, Tabular Editor.

📬 Find me at

Linkedin Badge Github Badge

💻 Tech Stack:

Agentic AI Large Language Models MCP AWS Bedrock Machine Learning Deep Learning Statistics MicrosoftSQLServer Azure Power Bi Bootstrap GitHub Pages Python HTML5 CSS3
✴️ Agentic AI ✴️ MCP ✴️ Machine Learning ✴️ Deep Learning ✴️ Statistics ✴️ Python ✴️ Azure ✴️ Azure DataFactory ✴️ Power BI ✴️ AWS ✴️ Data Modeling ✴️ Data Cleaning ✴️ Data Analysis
✴️ Debugging ✴️ Problem Solving

Pinned Loading

  1. Power-BI_Impact-of-Unemployment-on-Mental-Health Power-BI_Impact-of-Unemployment-on-Mental-Health Public

    This is my Power BI Project for the Master Program

    1

  2. Autonomous-MultiAgent-AWS-Incident-Jira-Automation Autonomous-MultiAgent-AWS-Incident-Jira-Automation Public

    AWS Incident Response System with LangGraph

    Jupyter Notebook 1

  3. MCP-Autonomous-Alpha-Agents MCP-Autonomous-Alpha-Agents Public

    Python 1

  4. ML-Predicting-Parkinsons-Disease ML-Predicting-Parkinsons-Disease Public

    Jupyter Notebook 1

  5. Project-DeepLearning-Human-Emotions-Detection-Model Project-DeepLearning-Human-Emotions-Detection-Model Public

    Jupyter Notebook 1

  6. Project-Fine-Tuned-LLMs-for-Product-Pricing Project-Fine-Tuned-LLMs-for-Product-Pricing Public

    Jupyter Notebook 1