shekharlabs
Shekhar Dudi

About

Shekhar Dudi

Lead AI Engineer building production-grade GenAI, agentic systems, and AI platforms that survive beyond the demo.

Melbourne, Australia Email LinkedIn GitHub

I’m an AI engineering leader with 13+ years of experience building scalable and production-grade software, machine learning, and Generative AI systems.

My work sits at the intersection of architecture, hands-on engineering, and technical leadership — turning ambitious AI ideas into systems people actually use, trust, and rely on.

I don’t just build AI demos. I build AI that works at 3 AM on a Tuesday.

Over the last decade, I’ve led and delivered solutions across enterprise virtual agents, RAG systems, multi-agent workflows, compliance automation, NLP products, MLOps platforms, and cloud-native AI infrastructure. My focus is on closing the gap between a promising prototype and a reliable product: strong retrieval, clear guardrails, structured outputs, observability, cost control, and engineering discipline from day one.

I’ve built and led AI teams from zero to one, partnered with executives and product teams, and mentored engineers to think beyond code — toward systems, outcomes, and long-term maintainability. My work has supported large-scale enterprise environments, including AI solutions serving 240,000+ users, reducing operational tickets by 40%, building search engines and multi-agent workflows.

How I think about AI

A demo proves that a model can answer.

A product proves that it can fail safely, recover gracefully, and keep delivering value.

That’s the bar I build for.

I care about AI systems that are useful, measurable, governed, and maintainable. That means designing with humans in the loop where needed, grounding outputs in the right data, tracing what happens under the hood, and making sure the system is cost-aware before it becomes expensive to operate.

What I build

I design and ship AI systems across:

Generative AI & Agentic Systems

LLMs, RAG, semantic search, multi-agent orchestration, tool use, prompt engineering, guardrails, and human-in-the-loop workflows.

AI Platforms & MLOps

Cloud-native AI platforms, CI/CD for ML systems, observability, evaluation pipelines, monitoring, cost controls, and scalable deployment patterns.

Architecture & Engineering Leadership

Technical strategy, solution architecture, engineering governance, stakeholder alignment, team mentoring, and translating business ambiguity into working systems.

Toolbox

Languages

PythonJava

AI & ML

LLMsRAGCrewAILangChainLangGraphLlamaIndexPyTorchTensorFlowscikit-learnNLP

Platform

AzureAWSKubernetesDockerTerraformKafkaOpenSearchPostgreSQLMongoDB

Off the clock

Outside the terminal, I’m usually planning the next trip, gaming, or tinkering with another side build that started as “just a quick idea.” I like building things — products, teams, systems, and occasionally overly ambitious travel itineraries.