Amar Chheda
Making AI Accessible!
Atlanta, Georgia, United States
Actions
Amar Chheda is an AI/ML engineer solving hard problems in the real world — from DeepFake detection and manufacturing vision systems to AI-driven cybersecurity. He brings deep technical expertise alongside a strong community ethos as the organiser behind AI Camp Atlanta, which has grown to over 4,000 members through hands-on meetups, hackathons, and industry panels.
Amar’s approach is refreshingly grounded: no hype, just clarity, execution, and impact. He’s spoken at conferences, mentored at hackathons, and actively collaborates with universities, tech communities, and industry leaders to make AI more accessible and actionable for professionals. Whether he’s building production-ready AI pipelines or helping others explore their first agent use case, Amar’s north star is always the same — using AI in service of real problems and real people.
Area of Expertise
Topics
Things I Learned While Building a Local-Only Agentic AI System: Superbrain
Coming from the pre-GPT world of data science and deep learning, I was used to AI systems being structured, constrained, and explicitly engineered. One trained statistical models, built CNN pipelines, measured performance, and controlled the behaviour through code.
My first exposure to LLMs and agentic AI felt almost alien: a markdown file, a long list of instructions, and a model expected to “figure out the rest.” To my old-school ML brain, that was both fascinating and deeply uncomfortable.
So I did what I know best: I built a system to learn how this new paradigm actually works.
In this session, I’ll share the lessons I learned while building Superbrain, a local-only agentic AI system designed to ingest information, organise knowledge, and answer questions with grounded citations, classify content, generate digests, and enable automated agentic flows, all while keeping data and inference private.
In this talk, I try to document my journey through making some architecture tradeoffs, engineering patterns and design decisions that made this possible.
I’ll cover:
- Why local LLMs are a great way to learn agentic AI.
- How smaller models force clarity in instructions, tighter workflow boundaries, and better system design.
- Why agentic AI is better understood as a set of orchestrated flows than a single autonomous “agent.”
- How retrieval, classification, summarisation, and follow-up actions combine into useful automated behaviours.
- What it takes to make these systems grounded, observable, and trustworthy.
I’ll also show how building with local models pushes you to think more carefully about:
- prompt and instruction design
- architecture boundaries
- tool contracts and workflow orchestration
- retrieval quality and groundedness
- logging, evaluation, and operational visibility
Attendees will leave with a practical mental model for building private, local-first agentic AI systems and a clearer understanding of how strong architecture matters even more when you are working with smaller local models.
Accelerating QA using Deep learning
In this talk, we will discuss the challenges of detecting and resolving performance issues, such as video stuttering, lag, and app crashes, and how Orange Bees has developed an AI-powered QA testing system to overcome these challenges. Combining Python, OpenCV, Object Detection, and OCR, the system validates video continuity and completeness, detects gaps, and identifies device-specific issues. With out-of-the-box report generation, businesses gain valuable insights into gaps and improve the app's user experience.
To validate the solution, we put it through rigorous testing using our own bench test. We recorded a clock on a desk for several hours using various phone devices and ran their analytics on top of that. The system was able to detect frozen frames and ensure that the recordings were complete and continuous, even when recording for extended periods. We also introduced artificially corrupted videos, introducing the exact issues that they were trying to detect, and the system was able to detect these issues perfectly.
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top