Open to opportunities

Sashikumar Nehrumohan

@sashikumarnehrumohan

Message

AI Engineer building production LLM pipelines, RAG, and agentic workflows at scale.

United States

Message

What I'm looking for

I’m looking for full-time AI Engineer roles where I can work on hard problems at scale—building production LLM/RAG systems, agentic workflows, and reliable MLOps pipelines that perform under real load and uncertainty.

I’m an AI Engineer with 3 years of production experience building LLM pipelines, RAG systems, and agentic workflows across financial services and enterprise AI. I’m a full-stack builder who can architect from scratch, debug production issues, and ship under ambiguity.

I specialize in multi-agent orchestration and RAG systems that handle real load—like an autonomous legal research agent built on LangGraph with Qdrant hybrid search, guardrails, and adversarial verification across 100+ concurrent sessions. I’ve also built a fine-tuned mental health voice agent using LoRA/PEFT, and delivered production impact such as cutting mortgage underwriting review time from 4 hours to 1 hour and reducing volunteer document verification from 2 hours to 15 minutes.

Experience

Work history, roles, and key accomplishments

Current

AI Engineer

Current

Rebecca Everlene Trust Company

Mar 2026 - Present (4 months)

*Architected a multi-agent pipeline utilizing LangChain orchestration and vector databases that automated volunteer onboarding processes, enhancing workflow efficiency by 40% within the first month of implementation. * Developed an intelligent document processing system that combined OCR technology with Python and LLMs to verify volunteer background documentation, reducing verification time from 2

Lang Chain LLM LLM Evaluation Multi Agent Systems Python Cloud Applications AI Security Embeddings NumPy Data Pipelines

Student Employee

University at Buffalo

Jan 2025 - Dec 2025 (11 months)

Worked as a student employee at the University at Buffalo for the 2025 academic year.

AI Engineer

Evu

May 2025 - Aug 2025 (3 months)

* Developed an LLM-powered GenAI document extraction system utilizing LLM and LangGraph that cut review time from four hours to just one hour per application, significantly increasing underwriter capacity.

* Achieved a 25% reduction in production costs through advanced techniques like context compression and prompt engineering while enhancing unit economics of mortgage document processing operat

Generative AI & LLM Applications (RAG Prompt Engineering LLM LLM Evaluation Agentic Automation Multi Agent Systems GuardRails Databricks Delta Lake Generative AI And LLM Applications (RAG CI CD Monitoring Embeddings Scalable AI System Design (End To End ML Production Deployment)

Machine Learning Engineer

Cognizant

Jun 2022 - Jun 2024 (2 years)

* Developed advanced Fraud and Risk prediction ML models on Databricks using PySpark and Delta Lake, decreasing runtime from 8 hours to just 5.6 hours while enabling rapid retraining for new fraud patterns.
* Built robust MLOps pipelines utilizing GitHub Actions, Docker, Kubernetes, and Argo CD for seamless model training and deployment processes that reduced release cycles from two weeks to one