At Olio Apps, we believe the success of every great product starts long before a single line of code is written.

It begins with discovery, a structured process that helps founders uncover the real problem their product solves, identify user needs, and separate must-have features from nice-to-have ideas.

In this video, we take you inside a real discovery session with Josh, who came to us with an exciting concept: a fitness app designed not just for individuals, but for group workouts.

You’ll see how we:

Peel back assumptions and refine the core problem.Identify what makes Josh’s idea unique (multi-user workout logging, group accountability, and goal-driven tracking).

Define what belongs in the minimum viable product (MVP) versus the roadmap.Explore the ergonomics and usability challenges of designing for the gym environment.

Discuss whether this prototype should be bootstrapped or positioned for investors.

By the end, you’ll understand why discovery sessions are so powerful for founders: they save time, reduce risk, and provide a clear, actionable roadmap.

Whether you’re an entrepreneur, innovator, or product leader, this session shows how slowing down early accelerates your path to market.

Work with us: If you have an idea for an app or product and want clarity before development, visit Olio Apps to schedule a discovery session.

If you found this video useful, don’t forget to like, subscribe, and share for more insights on product discovery, MVPs, and software innovation.

#FitnessApp #ProductDiscovery #MVP #Startup #AppDevelopment #OlioApps

Hey! We're Scott and Aron. We are software engineers, and business operators running Olio Apps. We help our clients build the best software possible. We're always learning and improving our methods and practices, and we want to share that with you.

--

Show Notes:

(0:00) Intro

(0:35) Welcome

(0:52) Zach’s Background

(2:22) Working with LLMs vs ML Models

(6:02) Applying AI to Backrs - Hyper Personalization

(9:01) From prototype to scale

(12:57) Streaming with Lambda

(17:34) Python streaming lambda wrapper

(22:17) Journey to get to 100 concurrent users - db connections - database pool via proxy

(25:21) Lifting LLM provider quota limits

(27:29) Lambda rate limits

(31:17) Backrs UI

(33:49) Backrs Streaming Example

(36:11) Conclusion

Hey! We're Scott and Aron. We are software engineers, and business operators running Olio Apps. We help our clients build the best software possible. We're always learning and improving our methods and practices, and we want to share that with you.

(00:00) – Intro & Use Case: Accelerating Prototyping with Gen AIDiscussing developer productivity needs in consulting environments and kicking off the demo.

(00:28) – Prompt-Based Prototyping Tools OverviewExploring tools like Bold.new, V0, and Replit that generate working apps from natural language prompts.

(01:14) – Message Board Prototype with Ratings & CommentsGenerating a message board app idea with features like image uploads, comments, and up/down voting.

(02:06) – Trying Classic Games: Snake and Pac-ManShifting from a complex message board to game prototyping as a test of tool capabilities.

(03:04) – Live Demo: Code, Database, and Hosting in BrowserWalking through how these tools scaffold code and backend infrastructure instantly.

(05:23) – Iteration Strategy: Adding Game Mechanics One Step at a TimeRealizing the importance of incremental prompting for better results in Gen AI.

(07:23) – Snake + Ghosts: Creativity in PrototypingMixing game ideas to create hybrid experiences and letting creativity drive prompts.

(09:22) – WebDevArena: Benchmarking LLMs in Real-TimeComparing outputs from models like Claude Sonnet and Gemini in a live coding battle.

(14:37) – E2B & the Future of App CustomizationDescribing E2B and predicting a future where AI builds user interfaces dynamically.

(16:07) – DiagramGPT: Serverless Architecture in SecondsUsing AI to generate AWS infrastructure diagrams and iterate with plain language prompts.

(22:33) – Schema Designer: Auto-Generating Database ModelsUsing AI to generate a SQL-based schema for a feature-rich message board.

(24:28) – Cursor IDE Tips: Rules, System Prompts & Model LayeringExploring how Cursor enhances coding with system prompts and layered model orchestration.

(29:28) - ComfyUI for Generative Video & Images

(33:46) - Closing Thoughts

In this episode, we cover:

00:00 – Agents vs. Workflows
What’s new in LLM-powered agents, and how they differ from traditional, on-rails workflows.

01:30 – Pattern Libraries and Agent Frameworks
Exploring tools like Amazon Bedrock and visualizing agentic workflows.

03:00 – LLM as Judge
Using higher-cost models to validate output from cheaper models. A clever way to balance cost and quality.

04:30 – Real-Time Evaluation and Prompt Engineering
How teams can move from static success criteria to live evaluation systems.

05:45 – System Prompt Design
Lessons learned from designing effective system prompts and using them the right way in production.

08:00 – Tool Use and LLM Decision Making
Teaching LLMs to use internal tools to answer queries—think calendar lookups, database queries, and more.

11:30 – End-to-End Testing with Natural Language
A new generation of testing libraries using LLMs and Playwright to turn plain English into functional tests.

12:45 – Linear vs. Branching Workflows in Data Pilot
How AI can define and adapt its own steps in data analysis and synthesis.

15:00 – Multi-Source LLM Querying
Merging inputs from time tracking, Slack, GitHub, Jira, and more to create richer outputs.

18:30 – Human-in-the-Loop Patterns
Designing workflows where humans review and approve AI-generated outputs before final delivery.

20:30 – Cursor and AI-Accelerated Development
How Olio engineers use Cursor for faster iteration while managing tech debt and consistency.

23:00 – Heuristics for Managing AI-Created Code
Tips for writing better system prompts, rules files, and maintaining consistent output across codebases.

27:00 – Infrastructure Throttling and LLM Scaling Challenges
Behind-the-scenes look at orchestrating LLM queries at scale, managing Lambda concurrency, and avoiding AWS throttles.

30:00 – LLM Identity & Masquerading Challenges
What it takes to make your AI chatbot actually believe it’s your brand—and how Claude and GPT-4 stack up.

33:00 – Looking Ahead to 2025
The team shares what they’re most excited to build next and how LLMs are becoming critical in everyday workflows.

Stay Connected
Subscribe to Code & Cognition for deep dives into AI, cloud-native development, and the tools shaping modern software teams.

Questions or feedback? Contact us at info@olioapps.com

Olio Apps can help you build your app: ⁠https://www.olioapps.com/contact-us⁠

Hey! We're Scott and Aron. We are software engineers, and business operators running Olio Apps. We help our clients build the best software possible. We're always learning and improving our methods and practices, and we want to share that with you.

From hallucinations and latency to the need for reliable structured output, we cover the hard lessons, practical tools, and architecture decisions that make the difference. You’ll also get a behind-the-scenes look at how Data Pilot helps teams prototype, test, and refine prompts using the latest LLMs like Claude and GPT-4.

If you're serious about shipping GenAI features at scale, this is the conversation for you.

Topics Covered:

1. The top 6 barriers to GenAI production adoption in 2024 and how best approach them

2. Why most LLM projects fail due to hallucinations and lack of guardrails

3. The role of system prompts and how to use them effectively

4. Debugging “self-talking” AI: how Claude can spiral without proper prompt roles

5. What structured output really means—and why it’s non-negotiable

6. Why prompt engineering is just the beginning—and data infrastructure is key

7. How Data Pilot streamlines prompt iteration, testing, and structured data generation

8. The hidden trade-offs between accuracy, latency, and cost when selecting LLMs

Tools & Concepts Mentioned:

1. System Prompts (LLM guardrails)

2. Retrieval-Augmented Generation (RAG)

3. JSON & XML output formats

4. Data Pilot – a workbench for LLM development

5. Claude, GPT-4, Meta’s LLMs

6. Prompt iteration and synthetic data generation

Key Themes:

"System prompts are the first line of defense—without them, the LLM just talks to itself."

"Structured output isn’t just nice to have—it’s how you scale, debug, and improve."

"Data Pilot lets you develop prompts with the help of an AI co-pilot that knows how it wants to be talked to."

Olio Apps can help you build your app: ⁠https://www.olioapps.com/contact-us⁠

Hey! We're Scott and Aron. We are software engineers, and business operators running Olio Apps. We help our clients build the best software possible. We're always learning and improving our methods and practices, and we want to share that with you.

Join us as Dustin, Engineering Manager at Olio Apps, walks through a real-world case study on optimizing Lambda concurrency for a growing SaaS product.

Learn how the team moved from a monolithic VPC setup to isolated environments, why “manual scaling” became unsustainable, and how they leveraged AWS Application Auto Scaling to provision smarter, not harder.

Whether you're just starting out with serverless or managing a mature product, this episode is packed with insights on modern infrastructure, developer productivity, and cost optimization in AWS.

Topics Covered:

• Why serverless architectures still need infrastructure planning

• Real-world lessons from working with a fast-scaling AWS customer

• The performance bottlenecks of Lambda cold starts

• Tradeoffs between provisioned concurrency vs. on-demand scaling

• Automating concurrency with AWS Application Auto Scaling

• Using CloudWatch metrics to fine-tune Lambda performance

• How isolated dev/staging/prod environments unlocked innovation

• The move from manual configurations to infrastructure as code

• Practical setup using the Serverless Framework and plugins

• Business wins: better performance, reduced costs, and fewer human errors

Key Takeaways:

• Lambda auto-scaling can drastically reduce costs while handling peak traffic efficiently.

• Investing in separate environments (dev, staging, prod) isn't just good practice — it enables risk-free innovation.

• Manual concurrency tweaks are a recipe for errors. Automation brings consistency and peace of mind.

• CloudWatch alarms and usage-based scaling let your infra grow with your users.

Resources

AWS Lambda Provisioned Concurrency

AWS Application Auto Scaling

Serverless Framework

CloudWatch Alarms

Olio Apps can help you build your app: https://www.olioapps.com/contact-us

Hey! We're Scott and Aron. We are software engineers, and business operators running Olio Apps. We help our clients build the best software possible. We're always learning and improving our methods and practices, and we want to share that with you.