Open to opportunities

Divyanshu Mittal

@divyanshumittal

Message

Data Scientist building production RAG and agentic systems with AWS-native MLOps.

India

Message

What I'm looking for

I’m looking for roles in LLM systems and agentic AI, plus applied ML infrastructure work. I want to own retrieval architecture, orchestration, and AWS-native MLOps—improving answer quality and end-to-end latency.

I’m a Data Scientist with 2 years of experience designing production RAG and agentic systems on AWS. I focus on retrieval architecture, LLM orchestration with LangGraph, and latency optimization, while shipping reliable, end-to-end pipelines using AWS-native MLOps.

At Adastra, I architected a production RAG pipeline for hierarchical patent retrieval—indexing 25,000 patents into OpenSearch with hybrid BM25 + FAISS-HNSW search—and made it the core retrieval layer for downstream client-facing research tools. I also integrated an Amazon Neptune knowledge graph for graph-aware pre-retrieval scoping and enrichment, built ingestion pipelines from USPTO and Google BigQuery into Athena-queryable formats with Glue/Step Functions/Terraform, optimized OCR extraction latency by 60%, and shipped Bedrock-based GenAI features with structured prompt templates and closed-loop feedback logging.

Experience

Work history, roles, and key accomplishments

Data Engineer Intern

Heart It Out

Aug 2024 - Oct 2024 (2 months)

Built and deployed 90+ n8n workflows to ingest data from web, app, and third-party sources into PostgreSQL and AWS storage layers, replacing manual data movement and improving analytics/data availability. Designed retry, monitoring, and alerting logic to reduce ingestion failures and manual intervention.

N8n Data Ingestion ETL PostgreSQL AWS S3 Workflow Orchestration Monitoring And Alerting Retry Logic

Education

Degrees, certifications, and relevant coursework

Indian Institute of Information Technology, Lucknow

Master of Science, Artificial Intelligence & Machine Learning

2023 - 2025

Grade: GPA: 9.06

Activities and societies: Class representative (CR) for the MSc (AI & ML) batch; participated in the company’s weekly tech talks.

Pursued an M.Sc. in Artificial Intelligence & Machine Learning, including research on federated learning for network traffic prediction in VANETs and a survey of ML techniques for intrusion detection in WSNs.