Miroslav Mitic
AI Engineer • Python Developer • Data Engineer
Software engineer with 10+ years of experience, focused on building AI-based applications, data platforms, and cloud-native systems. Specialized in LLMs, RAG architecture, AI evaluation frameworks, FastAPI, Python, AWS, Azure, and MLOps, with extensive experience leading engineering teams and delivering scalable solutions across the enterprise, finance, and technology sectors.
About
Building reliable AI systems, end to end
I'm an AI engineer and backend developer with a decade of experience spanning finance, enterprise, and technology. I focus on turning LLMs into dependable products — building evaluation pipelines, RAG architectures, and document-intelligence systems that hold up in production, backed by cloud-native infrastructure on AWS and Azure.
- 10+
- Years in software engineering
- 40%
- Predictive accuracy gained with RAG-based LLMs
- 35%
- Data-processing efficiency gained on AWS
- 99%
- Uptime on automated CI/CD deployments
Career
Experience
AI Engineer
Vitech · Vienna, VA
- Built an LLM evaluation pipeline for Gemini, Azure OpenAI, and GPT-4 using LangSmith, covering accuracy, schema compliance, and latency.
- Shipped a FastAPI LLM service with an OpenAI-compatible endpoint across mock and VertexAI Gemini backends.
- Automated model re-evaluation workflows and refined prompts from failure-case analysis, raising evaluation accuracy.
- Built an enterprise AI document parser (PDF, DOCX, PPTX, images) with hybrid OCR parsing via Azure Document Intelligence and PyMuPDF.
Backend Developer
Dell Technologies · Sunnyvale, CA
- Designed and optimized RESTful microservices, cutting response times by 30% under high traffic.
- Built Kafka-based event-driven pipelines for real-time data processing, mentoring 4 junior engineers along the way.
- Deployed Dockerized services on Kubernetes/AWS, increasing data-processing efficiency by 35%.
- Led RAG-based LLM predictive modeling, improving accuracy by 40% and cutting project turnaround by 30%.
Senior Python Developer
Wells Fargo · Charlotte, NC
- Built back-end services in Python, Django, and Tornado with RabbitMQ-integrated REST APIs.
- Built real-time AWS data pipelines with Spark/PySpark, increasing processing speed by 25%.
- Automated CI/CD with AWS CodePipeline, Jenkins, and Docker — 30% faster deploys, 99% uptime.
- Designed an AWS S3 data lake and deployed ML pipelines in Databricks for financial forecasting.
Software Developer
Idrica · Charlotte, NC
- Built secure REST APIs for IoT health-monitoring devices streaming real-time data on AWS ECS/Fargate.
- Applied machine learning to health data, improving predictive insights for operational efficiency.
- Optimized data workflows, increasing throughput for health-monitoring applications by 20%.
Toolbox
Skills
AI & LLM Engineering
Backend & APIs
Cloud & DevOps
Data Engineering
Background
Education & Languages
Washington State University
Master's Degree in Computer Science
2005 – 2008
Languages
- EnglishNative
- SpanishC2