Open to opportunities

James Johnson

@jamesjohnson92

Message

Senior AI Engineer building LLM-agnostic enterprise platforms with agentic RAG and production-grade ML impact.

India

Message

What I'm looking for

I want to build production enterprise AI—owning architecture for agentic RAG and Text-to-SQL, integrating secure privacy controls, and delivering measurable impact from deployment through iteration.

I’m a Senior AI Engineer focused on architecting production enterprise AI systems—especially LLM-agnostic platforms delivering Agentic RAG, multi-agent orchestration, and Text-to-SQL at scale. At 8bit.ai, I built semantic/keyword/hybrid enterprise search with Milvus and Elasticsearch, and engineered ReAct-style multi-agent workflows that enabled reasoning, query decomposition, tool use, and autonomous operations.

I also drive model and system performance through fine-tuning and efficient inference using vLLM and SGLang, deploying open-source models on GPU clusters within a model-agnostic architecture. I’ve delivered solutions to 4 enterprise customers, including chatbot-driven workflows built with Activepieces that reduced operational workload by 27% and improved analytics turnaround time.

Experience

Work history, roles, and key accomplishments

Current

Senior AI Engineer

Current

8bit.ai

Feb 2025 - Present (1 year 3 months)

Architected Neutrino, an LLM-agnostic enterprise AI platform delivering Agentic RAG, multi-agent orchestration, and Text-to-SQL at production scale. Delivered solutions to 4 enterprise customers, reducing operational workload by 27% and improving analytics turnaround time via chatbot-driven workflows.

Agentic RAG Multi agent orchestration Text To SQL Milvus Elasticsearch VLLM Sglang LLM Fine tuning Python

ML & Senior Software Eng

Threado

Aug 2022 - Jan 2025 (2 years 5 months)

Built production-grade PII detection and masking using DeBERTa-v3 with SIFT-based training and regex pipelines, enabling compliant AI workflows. Designed real-time ML pipelines with Kafka and CQRS, architected a production agentic RAG system, and deployed hybrid retrieval to support 50+ customers with secure document access via Notion OAuth.

DeBERTa V3 PII Masking Kafka CQRS Zero Shot Classification Agentic RAG Python

Research Assistant

Skyline University College

Jul 2019 - Jul 2022 (3 years)

Mentored 8 student programmers on software and machine learning projects and supported faculty with coursework development, evaluations, and academic program delivery.

Mentorship Machine Learning Support Student Evaluation

Machine Learning Intern

Inrobin Ltd

Oct 2018 - Dec 2018 (2 months)

Built a machine learning system for early equipment failure detection for an automotive manufacturer, achieving high prediction accuracy in production-like environments.

Predictive Maintenance Model Training data preprocessing Model Evaluation