Speaking / Talks · workshops · lectures
I speak and teach about trustworthy AI, agent security, Model Context Protocol systems, semantic data, energy-efficient machine learning, and human-centered AI-assisted development.
Upcoming keynote · October 8
· Event runs 6–9 p.m. ET. Introducing Pollard and the controls that let AI agents refuse, pause, and prove what happened before real-world actions. At Mana Tech Loft 10, 21 SE 1st Ave, Miami. The individual keynote time has not been published.
Event and registrationExplore PollardProgram-listed keynote
Date and invitation confirmation pending. The program lists this keynote, while the conference home omits it from current invited-speaker lists. Official pages also disagree on November or December dates, so this entry remains undated pending organizer confirmation.
Conference programConference home2026 engagements
Scheduled and delivered sessions verified against conference, organizer, or scheduling records.
A scheduled AiSalon Miami keynote on Pollard and governing AI agents before they act. The event runs 6–9 p.m. ET at Mana Tech Loft 10, 21 SE 1st Ave, Miami; the individual keynote time has not been published.
The accepted IEEE MLSP 2026 paper is scheduled for the 1–3 p.m. ET poster session in Atlanta, applying TOML’s physics-grounded energy model to 37 signal-processing algorithms.
Delivered to a standing-room-only audience at DeepStation AI Demo Night, Miami Dade College’s Wolfson Campus. A live workflow took a model through LoRA fine-tuning, ONNX INT8 optimization, evaluation, and release gates. Slides, measured results, and a runnable Colab exercise are available.
Demonstrated compact linked-data messages carrying trust and compliance annotations for constrained IoT devices at The DOCK in Miami.
Presented in the IEEE IRI 2026 Retrieval-Augmented Generation & Knowledge Graphs session in Seattle, covering temporal validity and stale-evidence controls for RAG.
Presented at IEEE IRI 2026 in Seattle, introducing uncertainty-aware guardrails for LLM-driven IoT actuation.
A hands-on GDG Central Florida workshop using JSON-LD Ex to represent confidence, uncertainty, provenance, and compliance in AI systems.
An AgentCon Miami main-track talk and interactive defense lab on MCP, agent, and AI-pipeline attack surfaces.
Four scheduled paper and poster sessions in Marco Island, including oral presentations on JSON-LD Ex and wearable neck-posture classification.
An Orlando Code Camp session on extending linked data with AI confidence, provenance, temporal scope, and security metadata.
2025 engagements
Public instruction across agentic AI, fine-tuning, model evaluation, and custom GPTs.
A hands-on technical masterclass at The DOCK in Miami covering neural networks and model fine-tuning.
A Vibe Days session at The LAB Miami on rapid interface building and prompt-driven development.
A DeepStation and GPTechday workshop in Miami on retrieval, safety controls, and mixture-of-experts techniques.
OpenAI Academy lists Muntaser Syed as a speaker for this online workshop, presented by DeepStation and co-presented with Grant Kurz.
A South Florida AI Summit session at The LAB Miami on separating credible AGI trajectories from marketing claims.
Earlier appearances
Earlier talks, program listings, and accepted presentations spanning robotics, UAVs, cybersecurity, and the hacker community.
A 55-minute guest interview on The Hackathon Podcast.
A South Florida GIS Expo presentation in West Palm Beach with Gamal Bohouta and Veton Z. Këpuska.
Accepted poster for NAECON 2018, with Muntaser Syed registered as presenting author. Proceedings publication is not established.
A FLAIRS-31 poster presentation in Melbourne, Florida.
A Florida Conference on Recent Advances in Robotics presentation at UCF.
Accepted abstract by Muntaser Syed and Mohammad Algarni, listed in AHFE 2016’s Human Factors in Cybersecurity session. The program listing does not establish a published full paper or confirm delivery.
Teaching archive
Labs, workshop curricula, and session pages that remain useful beyond a single event.
Slides, a complete Colab exercise, measured results, and an attendee handout for fine-tuning, optimizing, evaluating, and releasing a model.
A one-hour lecture and two-hour hands-on workshop covering gateways, production architecture, and advanced MCP implementations.
An interactive lab on MCP, autonomous-agent, and AI-pipeline attack surfaces with defensive patterns.
Notebooks and code samples labeled for an ACM lecture on agent architectures and practical implementation patterns; no public organizer page was located.
Current abstracts and session formats for trustworthy AI, MCP security, JSON-LD Ex, XAI evaluation, and human-centered development.
Invite / collaborate
For conference sessions, workshops, guest lectures, or technical panels, email me with the audience, format, and proposed date.