📍 London, United Kingdom
I am an AI Engineer passionate about transforming Generative AI ideas into production-ready solutions.
My experience focuses on designing and deploying AI applications, multi-agent systems, RAG architectures, and LLM-powered automation solutions using AWS cloud technologies. I specialize in building reliable, scalable AI systems with strong emphasis on evaluation, guardrails, observability, and MLOps best practices.
- AWS Certified Machine Learning – Associate (MLA-C01)
- AWS Certified Data Engineer – Associate (DEA-C01)
- AWS Certified AI Practitioner
- AWS Certified Cloud Practitioner (CLF-C02)
- Microsoft Power BI Data Analyst Associate (PL-300)
- Amazon Bedrock
- Claude Models (Sonnet / Opus)
- LangGraph
- LangChain
- LangSmith
- LlamaIndex
- Agentic Workflows
- Multi-Agent Systems
- Prompt Engineering
- AI Evaluation Frameworks
- LLMOps
- AI Guardrails
- Vector Databases (FAISS)
- Semantic Search
- Embedding Models
- Amazon Titan Embeddings
- Knowledge Bases
- Document Intelligence
- Context-Aware Retrieval
- Scikit-learn
- TensorFlow
- PyTorch
- Transformers
- Random Forest
- XGBoost
- CNN
- RNN
- LSTM
- BERT
- ALBERT
- Hugging Face
- AWS Bedrock
- AWS SageMaker
- AWS Lambda
- AWS ECS
- AWS ECR
- AWS S3
- AWS Glue
- AWS Athena
- AWS CloudWatch
- AWS IAM
- Docker
- GitHub Actions
- CI/CD Pipelines
- Python
- FastAPI
- Streamlit
- REST APIs
- Pydantic
- SQL
- Power BI
- DAX
- Pandas
- NumPy
- Data Modelling
- ETL Pipelines
- Dashboard Development
- Built production-grade multi-agent AI systems using LangGraph and Amazon Bedrock.
- Developed RAG pipelines combining Bedrock, LangChain, FAISS, and semantic retrieval.
- Automated document generation workflows with up to 98% accuracy.
- Designed AI-powered invoice extraction agents integrated with ERP systems.
- Implemented LLM evaluation frameworks using golden datasets and LLM-as-a-judge methodologies.
- Built deployment pipelines using Docker, AWS ECS/ECR, and GitHub Actions.
- Established observability using LangSmith and AWS CloudWatch.
- Developed NLP classification systems using ALBERT, Random Forest, XGBoost, and LSTM models.
- Built AWS-based data pipelines integrating EC2, S3, Glue, and Athena.
- Delivered business intelligence dashboards and stakeholder reporting using Power BI.
- Improved model quality through advanced data labeling and class balancing techniques.
Production AI platform that transforms meeting transcripts and business documents into structured Proof of Concept (POC) documents using Amazon Bedrock, LangGraph, LangChain, and FAISS.
Tech Stack
- AWS Bedrock
- Claude Models
- LangGraph
- LangChain
- FAISS
- Python
- AWS Lambda
- S3
AI-powered invoice extraction and validation system that automatically processes invoices and posts structured outputs to ERP systems.
Features
- LLM-based Extraction
- Pydantic Guardrails
- Automated Validation
- CI/CD Evaluation Gates
- ERP Integration
Agentic AI solution capable of handling real-world travel planning queries using AWS Bedrock Agents.
Features
- Multi-step Reasoning
- Tool Calling
- Response Evaluation
- Confidence Scoring
- Production Monitoring
Advanced text classification solution using ALBERT embeddings, Random Forest, XGBoost, and LSTM architectures.
Achievements
- 400+ manually labelled records
- Class imbalance mitigation
- 80% classification accuracy
- Automated label generation pipeline
End-to-end deployment framework for AI applications.
Components
- Docker
- AWS ECS
- AWS ECR
- GitHub Actions
- CloudWatch
- LangSmith
- CI/CD Automation
Middlesex University, London
- GitHub
- Email: negarbaibordiii@gmail.com
Always interested in discussing:
- Generative AI
- Agentic AI
- AWS Bedrock
- LLMOps
- MLOps
- Production AI Systems