Just Now Possible

What does it take to build an AI that can take a food order over WhatsApp — correctly, every time, fast enough that customers can't tell it's not a person? That's the core challenge Santi Marchiori and Juan Haedo set out to solve at AITropos, a company building AI employees for the hospitality industry. In this episode of Just Now Possible, Teresa Torres talks with Santi Marchiori (CEO) and Juan Haedo (CTO) of AITropos about how they built an AI order-taking agent that handles the full flow — menu recommendations, modifiers, delivery zones, payment links, and status updates — entirely inside WhatsApp. They went through three product iterations to get there: first a hardware device for waiters, then a waiter-facing app, and finally a customer-facing conversational agent powered by a tools-based architecture designed for speed and reliability. You'll hear how they solved the core technical challenge of translating non-deterministic human conversation into structured POS-compatible order data, why they chose tools over MCP for agent architecture, how they pre-inject product context to cut latency before the agent ever makes a tool call, and why they test with thousands of agent-simulated customer conversations overnight before deploying to any real venue.

What if the future of product design isn't about replacing designers — it's about giving every team access to one? For solo founders, stretched design teams, and early-stage startups, great UX has always been out of reach. Banani is trying to change that by building an AI product designer that doesn't just generate code — it generates design. In this episode of _Just Now Possible_, Teresa Torres talks with Vlad Solomakha (CEO & Co-founder), Vova Parkhomchuk (CTO & Co-founder), and Vlad Ostapovats (Founding Growth) about how they built Banani from a Figma plugin proof-of-concept into a canvas-first AI design tool generating hundreds of thousands of designs per week. Vlad Solomakha brings a decade of design experience to the product — and a very specific vision of what it means for AI to produce beautiful, tasteful design rather than average, undifferentiated UI. You'll hear how they engineered their agent to handle parallel screen edits, manage per-screen context across canvases with hundreds of frames, and make surgical edits without regenerating entire screens. They dig into the "gulf of specification" — the mismatch between how designers think visually and how agents understand text — and what they're building to close it. It's a detailed look at what it takes to build an AI-native design tool that puts the designer in the driver's seat while letting the agent handle the production work.

What if an AI could spot the moment two product teams start pulling in opposite directions -- before it derails a quarter? In this episode of Just Now Possible, Teresa Torres talks with Matthias and Charlotte Kleverud, co-founders of Momental, about their vision for building what they call "GitHub for product management." Momental ingests documents, meeting transcripts, and voice recordings across an organization, then uses AI agents to map them into a structured context layer—a set of interconnected trees covering goals, decisions, learnings, and who's doing what. When it finds a conflict—say, one team betting on retention while another is prioritizing conversion—it surfaces the misalignment for humans to resolve, just like a merge conflict in code. You'll hear how their journey started in 2022 with GPT-3, pivoted through a multi-agent team that exposed the real problem (agents need aligned context, too), and landed on an OODA-loop-driven document processing agent that builds and maintains a living knowledge graph. It's a detailed look at how domain-specific context modeling, human-in-the-loop conflict resolution, and self-improving agents come together to tackle one of the hardest problems in product organizations: keeping everyone aligned.

What if your meetings could actually produce the artifacts you need—specs, tickets, slides—before the call even ends? In this episode of Just Now Possible, Teresa Torres talks with Mark Barbir (CEO) and Sanden Gocka (Co-Founder), the co-founders of Earmark, about building a productivity suite that turns unstructured conversations into finished work in real time. Unlike generic AI notetakers that produce summaries nobody reads, Earmark runs multiple agents in parallel during your meetings—translating engineering jargon, drafting product specs, even spinning up prototypes in Cursor or V0 while you're still talking. You'll hear how they pivoted from an Apple Vision Pro presentation coaching tool to a web-based meeting assistant, why their ephemeral (no-storage) architecture became a feature for enterprise sales, and the technical challenges of making real-time AI affordable—from $70 per meeting down to under a dollar through prompt caching. They also dig into why vector search falls short for analysis questions and how they're building agentic search to find insights across months of meetings. Whether you're a PM drowning in follow-up work or a builder curious about real-time AI architectures, this conversation offers a detailed look at what it takes to ship an AI product that people can't imagine working without.

When a construction company receives a bid request, someone has to open that email, parse the attached PDF (sometimes 1,800 pages describing an entire building), figure out which products are relevant, look up pricing, and draft a quote—all before the deadline. It's tedious, error-prone, and surprisingly manual. In this episode of _Just Now Possible_, Teresa Torres talks with Daniel Kappler (CTO, Product & Design) and Matthias Hilscher (CTO, Engineering) from Tendos AI about how they're automating this entire workflow for manufacturers in the construction industry. What started as a narrow prototype matching radiator requests to product catalogs has grown into a full agentic system that handles everything from email categorization to offer generation. You'll hear how they validated the opportunity with a design partner, spent a week on-site watching users work, and built a multi-agent architecture where specialized agents collaborate—complete with a "review agent" that checks the work of other agents before anything reaches a human. They dig into why they evaluate each agent independently (not just the whole chain), why they built custom observability tools when off-the-shelf solutions fell short, and how human-in-the-loop feedback is pushing them toward a self-learning system.

What happens when a customer reports a stolen credit card? The frontline answer is simple—freeze it. But underneath lies a cascade of follow-ups: dispute filings, fraud investigations, merchant communications, and proactive outreach to gather more details. Most AI support tools handle only the tip of the iceberg. In this episode, Teresa Torres talks with Jack Taylor (Product Engineer) and Ibrahim Faruqi (AI Engineer) from Gradient Labs, an AI-native startup building agents that automate the full scope of customer support in fintech. They share how they've architected a platform with three coordinating agents—inbound, back office, and outbound—all built on a shared foundation of natural language procedures, modular skills, and configurable guardrails. You'll hear how they: - Let non-technical subject matter experts define agent behavior through natural language procedures—no coding required - Architected a state machine orchestrator that manages turns, triggers, and skill selection across long-running conversations - Built guardrails as binary classifiers with eval pipelines, tuning for high recall on critical regulatory checks - Designed an auto-eval system that samples conversations for human review to catch edge cases and build labeled datasets It's a detailed look at how one startup is moving beyond simple Q&A bots to agents that can actually take action, coordinate across workflows, and handle the messy reality of customer support.

What happens when you combine a real customer problem, a no-code prototype, and a team willing to listen to every single call? In this episode of _Just Now Possible_, Teresa Torres talks with Steven Payne (Product Manager), Gabriel Stock (Senior Engineering Manager), and Philipe Steiff (Senior Software Engineer) from Perk—a company that helps businesses eliminate "shadow work" like travel booking and expense management. They share how they built a voice AI agent that calls hotels to verify virtual credit card payments, preventing travelers from arriving to find their rooms unpaid. What started as a hackathon experiment in Make.com became a production system handling over 10,000 calls per week across multiple languages. Along the way, the team learned hard lessons about prompt engineering for voice (numbers, pronunciation, and a very "Karen-like" first version), how to break a single monolithic prompt into structured conversation stages, and why listening to actual calls beats any amount of theorizing. You'll hear how they: - Built a working prototype without writing a single line of backend code - Structured the call into discrete stages (IVR, booking confirmation, payment) to improve reliability - Created two eval systems: one for call success classification, another for conversational behavior - Scaled from five calls a day to tens of thousands per week while maintaining quality This is a detailed look at building AI for real-time human interaction—where the stakes are high and the feedback is immediate.

Accounts payable inboxes can see 1,000+ vendor emails a day. Xelix's new Helpdesk turns that chaos into structured tickets, enriched with ERP data, and pre-drafted replies—complete with confidence scores. In this episode, Claire Smid (AI Engineer), Emilija Gransaull (Back-End Tech Lead), and **Talal A.** (Product Manager) walk us through how they scoped the problem, prototyped with “daily slices” (Carpaccio-style), and built a retrieval-first pipeline that matches vendors, links invoices, and drafts accurate responses—before a human ever clicks “send.” We dig into tricky bits like vendor identity matching, Outlook threading, UX pivots from “inbox clone” to ticket-first views, and the metrics that prove real impact (handling time, stickiness, auto-closed spam). We close with what’s next: targeted generation, multiple specialized responders, and more agentic routing.

Trainline—the world’s leading rail and coach platform—helps millions of travelers get from point A to point B. Now, they’re using AI to make every step of the journey smoother. In this episode, Teresa Torres talks with David Eason (Principal Product Manager) Billie Bradley (Product Manager), and Matt Farrelly (Head of AI and Machine Learning) from Trainline about how they built Travel Assistant, an AI-powered travel companion that helps customers navigate disruptions, find real-time answers, and travel with confidence. They share how they: - Identified underserved traveler needs beyond ticketing - Built a fully agentic system from day one, combining orchestration, tools, and reasoning loops - Designed layered guardrails for safety, grounding, and human handoff - Expanded from 450 to 700,000 curated pages of information for retrieval - Developed LLM-as-judge evals and a custom user context simulator to measure quality in real-time - Balanced latency, UX, and reliability to make AI assistance feel trustworthy on the go It’s a behind-the-scenes look at how an established company is embracing new AI architectures to serve customers at scale.

What if your next teammate was an AI coworker — one that could answer support tickets, process invoices, or even draft your next email — and your _non-technical_ colleagues could teach it how to do those tasks themselves? In this episode, host Teresa Torres talks with Seyna Diop (CPO), Job Nijenhuis (CTO & Co-founder), and Christos C. (Lead Design Engineer) of Neople, a company creating “digital coworkers” that blend the reliability of automation with the empathy and flexibility of AI. They share how Neople evolved from simple response suggestions to fully autonomous customer service agents, the architecture that powers their conversational workflow builder, and how they designed eval loops that include their _customers_ as part of the quality process. You’ll learn how the team: - Moved from “LLMs will solve everything” to finding the right balance between code, agents, and guardrails - Designed evals that run in production to detect hallucinations before an email ever reaches a customer - Helped non-technical users build automations conversationally — and taught them decomposition along the way - Turned customers’ feedback loops into eval pipelines that improve product quality over time It’s a fascinating look at how one startup is rethinking what it means to “work with AI” — not as a tool, but as a teammate.

How do you know if your AI product is actually any good? Hamel Husain has been answering that question for over 25 years. As a former machine learning engineer and data scientist at Airbnb and GitHub (where he worked on research that paved the way for GitHub Copilot), Hamel has spent his career helping teams debug, measure, and systematically improve complex systems. In this episode, Hamel joins Teresa Torres to break down the craft of error analysis and evaluation for AI products. Together, they trace his journey from forecasting guest lifetime value at Airbnb to consulting with startups like Nurture Boss, an AI-native assistant for apartment complexes. Along the way, they dive into: - Why debugging AI starts with thinking like a scientist - How data leakage undermines models (and how to spot it) - Using synthetic data to stress-test failure modes - When to rely on code-based assertions vs. LLM-as-judge evals - Why your CI/CD set should always include broken cases - How to prioritize failure modes without drowning in them Whether you’re a product manager, engineer, or designer, this conversation offers practical, grounded strategies for making your AI features more reliable—and for staying sane while you do it.

When ChatGPT launched, Stack Overflow faced a cataclysmic shift: developer behavior was changing overnight. In this episode, Teresa Torres talks with Ellen Brandenburger, former product leader at Stack Overflow, about how her team navigated the disruption, prototyped AI features, and eventually built an entirely new business line. Ellen shares the inside story of Overflow AI—from the first scrappy prototypes of conversational search, through multiple iterations with semantic search and RAG, to the tough decision to roll the product back when it couldn’t meet developer standards. She also explains how Stack Overflow turned a looming threat into opportunity by creating technical benchmarks and licensing its Q&A corpus to AI labs. This episode offers a rare look at what it really takes to adapt when a platform-defining shift hits—and what product managers, designers, and engineers can learn about prototyping, evaluating quality, and building in uncertainty.