🛠️ Why Voice AI Agents Fail: A Field Guide

On May 26th, we are gathering Chicago’s most active engineers, founders, and operators at Drive Capital to discuss the engineering reality of voice AI. While voice agents in 2026 often demo beautifully, they frequently fail in production due to complex turn-taking, state management, and the challenge of separating human signal from noise.

Most voice agent failures are not model failures; they are failures in the orchestration layer. This session features Rod Rivera from Rasa, who will walk through four specific production failures, demonstrate them with running code, and share an open-source repository to help you reproduce and solve these challenges in your own stack.

🗣️ Featured Speaker

Rod Rivera, Developer Relations at Rasa. With 15+ years in applied ML, Rod focuses on the practical engineering required to move beyond “audio-glued-to-text” and into resilient, production-grade voice interfaces.

🎙️ Show Your Work: Call for Demos

We are looking for active builders to present technical deep-dives or demos. We prioritize code over slides and engineering over sales pitches. If you have solved complex state management challenges or built novel agentic workflows, we want to see it.

Submit Your Demo Proposal Here

🗓️ Schedule and Logistics

• Date: Tuesday, May 26, 2026
• Time: 5:30 PM - 8:30 PM
• Location: Drive Capital, Chicago (Exact address provided upon approval)
• Format: Technical deep-dive followed by networking and community demos.

🛡️ Curation Policy

Attendance is strictly limited to 100 practitioners. We screen every registration to ensure the room is filled with engineers and researchers who are actively shipping. Space is limited and our Chicago events consistently run with a waitlist.

🤝 Our Partners

Drive Capital is a venture capital firm managing $2.2B in assets, partnering with founders across America to build market-defining companies from seed through IPO.

Rasa provides the infrastructure for enterprises to build and deploy sophisticated, mission-critical AI agents that users can trust.

Event photos

AI Tinkerers - Chicago

AI Tinkerers - Chicago

AI Tinkerers - Chicago

AI Tinkerers - Chicago

AI Tinkerers - Chicago

AI Tinkerers - Chicago

📊 AI Tinkerers Chicago Stats

• Attendees: An exclusive network of 2,167 AI professionals, this community consists of 88% machine learning specialists, 74% software engineers, and 58% cloud infrastructure experts. Notable achievements include members pioneering production-grade agentic workflows, deploying custom LLM pipelines in regulated healthcare and fintech sectors, and spin-outs securing venture backing from premier global accelerators.
• Companies Represented: Featuring tech giants like Google, Microsoft, and NVIDIA, alongside fast-growing platforms such as DoorDash, TikTok, and GitLab, and innovative startups like ElevenLabs, Pinecone, Speechify, Baseten, and HumanLayer, and more
• Demos: 182 demos have been submitted and 107 have been presented. The most exciting themes have focused on agentic systems (often coordinated via tool calling and structured handoffs), robust retrieval-driven applications (including graph/vector/RAG variants), and production-minded reliability (schema enforcement, guardrails, and eval/observability loops). Technical highlights have also included multimodal pipelines (OCR, voice, image-to-structure) and cost/performance optimization for real deployment.
• Testimonials:

  “...Events like this are very helpful for students and professionals looking to grow their technical skills and better understand the future of AI and technology.”

  — QA engineer, cloud support

  “...Your network puts you in a unique position to really make this idea scale, and reach the right people.”

  — Founder, crypto AI

  “Continue keeping things technical ...”

  — AI engineer, fintech

  “Please continue organizing highly technical, builder-focused AI events like this in Chicago.”

  — R&D engineer, defense tech

  “A beautiful and cozy space -- really perfect for this type of event!”

  — Cybersecurity entrepreneur

  “Keep the community going it’s a great group.”

  — Platform Engineering Lead

  “Keep providing a down-to-earth atmosphere for founders and developers to connect.”

  — Entrepreneur, AI/ML

A great demo, per the strongest examples, feels like a working technical artifact plus a teachable pattern: it should solve a specific engineering challenge, show the mechanism (ideally with inspectable intermediate artifacts like structured outputs, logs, orchestration graphs, or “contracts” between components), and provide reusable lessons/trade-offs other builders can apply. This aligns with the speaker-form requirement to avoid slide decks/marketing pitches and instead demonstrate working software, messy experiments, or architectural walkthroughs. Demos do well when they make “correctness/robustness” or “human agency” tangible through the system’s structure (e.g., building safer behavior into the architecture, separating responsibilities like abstention/routing/approval, and providing a durable interface where the human can review and edit rather than relying on vague prompt instructions). Conversely, avoid demos that read like landing pages, have missing/placeholder content, or don’t clearly show the core implementation details and what audience members can observe directly—especially if the main takeaways are generic or marketing-oriented rather than concrete and builder-focused.

In Chicago, Pat Narendra’s Agentic Video Studio: Building a Human-Editable AI Video Synthesis System stood out for its agent workflow where output becomes inspectable/editable project state rather than a one-shot “AI made a video.” The key reason it’s compelling is the explicit contracts (scene-first manifests, human approval surfaces, and deterministic compilation), which makes the workflow feel like engineering instead of magic. Also in Chicago, Luis Cisneros’ Teaching a Clinical Multi-Agent System to Say I Don’t Know earned a perfect score in the provided data by demonstrating abstention and calibrated uncertainty in a clinically grounded multi-agent setup; it’s engaging because the system surfaces interpretable disagreement and formally routes cases outside its tolerance rather than averaging everything into confident text. Finally, James Meyer’s 5 Human Sparks - What is work in the AI future? scored 5/5 by reframing the human-AI relationship with a clear framework (Judgement, Presence, Empathy, Creativity, Purpose), addressing audience anxiety about AI replacing work by shifting from hype to a design philosophy that builders can apply when deciding what should remain human.