Kiriti & Aish

Kiriti & Aish

43.3K Subscribers

AI Founder & Advisor to F500s | Ex-AWS

By continuing, you agree to Maven's Terms and Privacy Policy.

Kiriti & Aish will make you a problem-first AI builder

Aish Reganti is the founder and CEO of LevelUp Labs, a human-first AI startup that helps develop and deploy production-ready generative AI solutions for mid to large enterprise clients. With over 10 years of experience in machine learning, she has published more than 40 research papers at top-tier AI conferences, including NeurIPS, AAAI, and CVPR.

Kiriti Badam is a member of the technical staff at OpenAI Codex, with over a decade of experience designing high-impact enterprise AI systems. He specializes in AI-centric infrastructure, with deep expertise in large-scale compute, data engineering, and storage systems.

Worked/Taught At
MIT
University of Oxford
Amazon Web Services
Microsoft

Alumni reviews

Ricardo

Live cohort
I came in as a Senior PM who had designed agentic systems but never built one end to end. This course closed that gap What I valued most was how complete it was. Instead of jumping straight to agents, it covered the whole system around Agents: problem framing, context engineering, RAG and memory, MCP and skills, workflows vs. autonomous and multi-agent designs, evals, cost and latency, fine-tuning, and multimodal. Each topic connected back to one mental model, so it never felt like a list of tools It's also genuinely hands-on. Over five weeks I built a router and a ReAct agent, hybrid RAG with a proper eval set, multi-agent deep research with memory over MCP, and a small-vs-frontier model comparison Above all, it's honest. In a space full of hype, the course kept coming back to the simplest solution first, evaluating before adding complexity, and reasoning about trade-offs. One tip for future students: the pace is intense, so plan time to revisit the material afterwards. It's worth it.

Vijay

Live cohort
An excellent, well-structured course that goes beyond theory and focuses on practical, hands-on learning. The iterative coding assignments, capstone project, live sessions, and highly engaged instructors make the experience especially valuable. I particularly appreciated the accessibility of the instructors, responsive Slack support, and the willingness to adapt the content based on student feedback. A great combination of depth, practical application, and community. Highly recommended!

Prateek

Live cohort
Product and Startegy Leader · Independent

Building Agentic AI Applications with a Problem-First Approach

Aishwarya and Kiriti have built one of the most substantive AI courses on Maven. 'Building Agentic AI Applications with a Problem-First Approach' goes well beyond prompting - it covers the full engineering stack behind production AI systems, with real emphasis on designing them well. They make complex concepts accessible without dumbing them down. As someone non-technical, the few times I did feel out of my depth, the Grow lessons had extra content to build the basics and then go even deeper. Weekly assignments make you apply everything hands-on, and between office hours, Chai & AI, Build Labs and an always-open Slack, help is never far. I came away with a proper systems view of AI - what it can do, where its limits are, and how to design around both. Highly recommend.

Gagan

Live cohort
This course is a rollercoaster ride. The instructors have packed layers of learning week after week, which keeps your mind engaged. In the end, I am confident that I can decipher the AI conversations in meeting rooms and the technical videos I watch on YouTube. All the topics line up, giving me that 'aha' moment that this course has taught me the exact fundamentals needed for this expanding AI world.

Graham

Live cohort
The content was extremely thought out and well delivered. The practical assignments extended the learning by forcing us to think for ourselves to accomplish tasks and giving us broad familiarity to different AI tools. I particularly appreciated that this course focused on the framework and scaffolding to make sense of modern AI tools rather than overly focusing on specific tools or techniques that could become quickly dated.