Instructors

Taught by Engineers Who Build AI Systems, Not Just Explain Them

Every instructor on the Generative AI & AI Application Development course works — or has worked — on real production AI and software systems. That's a deliberate choice: the curriculum is shaped by what actually breaks in production, not just what looks good in a slide deck.

Nishi Singh

Research Scientist, PhD

LLM evaluationResearch methodologyModel behaviour & hallucinations

Nishi holds a PhD and works as a research scientist, where her day-to-day involves designing experiments, validating hypotheses with data, and holding ideas to a standard of evidence before they ship.

Inside the Synaptech classroom, she brings that same rigour to how generative AI is taught: not "prompt and hope", but a methodical way to evaluate whether an LLM output, a RAG pipeline, or an agent is actually working — and how to prove it with numbers instead of vibes.

She leads the sessions on model behaviour, hallucination analysis, and LLM evaluation, and reviews the course’s technical content for accuracy before it reaches students.

Satish Kumar

Software Development Manager, Amazon

LLM app developmentProduction engineeringScalable backend architecture

Satish manages software development teams at Amazon and has shipped high-scale applied AI automation used well beyond a single team or product.

That background shapes how he teaches: less "toy demo", more "what breaks in production and why" — rate limits, retries, cost per request, and the engineering discipline it takes to keep an AI feature running reliably once real users show up.

He teaches the LLM application development and production AI modules, where students move from a script that calls an API to a backend that can actually be deployed.

Pravesh Dholwani

Applied AI Engineer

Retrieval-Augmented GenerationAI agents & tool callingApplied AI infrastructure

Pravesh is an applied AI engineer who has scaled AI and backend infrastructure for products used by 10 million users — the kind of scale where a shortcut in the architecture shows up as a real outage, not a warning in a console.

He designed Synaptech’s curriculum around building, not just watching: every concept — tokens, embeddings, RAG, agents — is paired with a working project the same week it’s taught, because that’s how he learned to ship AI systems himself.

He leads the RAG & knowledge systems, AI agents, and capstone project modules, and is the course’s primary instructor for hands-on build sessions.

Nitish Kumar

AI Team Lead, ISRO

Multimodal AIApplied AI engineeringAI reliability

Nitish leads the AI team at the Indian Space Research Organisation (ISRO), where the acceptable margin for error in an AI system is close to zero.

He brings that discipline to how Synaptech students are taught to think about multimodal AI, tool-calling reliability, and what it actually takes for an AI application to be trusted with a real task.

He co-teaches the AI agents, multimodal AI, and production-readiness sessions, and is a regular judge for the course’s in-class "prompt battle" and demo day.

Why it matters

Instructor quality is a curriculum decision, not a marketing line

A generative AI course is only as current as the people teaching it. Our instructors actively work on applied AI, backend systems at scale, and AI research — so the curriculum gets updated when the field moves, not once a year.