In this episode of Engineering Evolved, Tom sits down with Amelia Prasad, Director of Product at Concept to Cloud, to trace how AI has reshaped the day-to-day of UX and product design. Amelia — who came to product from astrophysics and climate science — walks through the shift from manual research, whiteboards and "back-of-the-napkin" sketches to building zero-to-one products directly in Claude Code.

It's not a hype reel. Amelia is candid about the friction: learning version control from scratch, bloated six-thousand-line files, the designer-to-developer handoff problem, and the diminishing returns of heavy token usage in tools like Claude Design and third-party wrappers such as Lovable. Her current answer is a marriage of tools — passing work back and forth between Claude Code and Figma via MCP — so prototyping speed and real usability, accessibility and design-system rigour can each live where they belong.

They close on what the next 12 months might hold: more human-led user research, not less, and why juniors and design intuition still matter in an industry tempted to hire only senior builders.

Chapters

• 00:00 — Welcome & introducing Amelia
• 02:53 — From astrophysics to product design
• 03:33 — The pre-LLM UX workflow: manual research & competitive analysis
• 05:23 — Old-school tooling: Figma, Miro, Maze, pen & paper
• 06:56 — The lost art of napkin sketches and paper prototyping
• 07:57 — Meeting at Princeton: first exposure to LLMs
• 08:52 — AI workflows before AI building: the interview note-taker
• 11:02 — Stepping into Claude Code as a non-developer
• 13:45 — Handing off code on a small team: value and limits
• 15:17 — Guardrails, context and the handoff problem
• 20:02 — Lovable, Cursor and the trouble with wrappers
• 21:05 — Claude Design: token cost and diminishing returns
• 22:45 — Figma's MCP and the two-way handoff
• 24:50 — A suite of tools: knowing when to hand off to which
• 29:06 — Why active engagement in Figma beats waiting on the terminal
• 33:02 — The next 12 months: user research, systemic processes, robustness
• 38:16 — Will design jobs disappear? Juniors, intuition & human-in-the-loop
• 41:27 — Wrap-up & thanks

Key takeaways

• AI adoption in design was gradual — automating admin and research before it could build whole products.
• The real skill now is orchestration: knowing which tool (Claude Code vs. Figma) does which job best.
• Speed doesn't replace craft — usability, accessibility and design systems still need a designer's hand.
• Human intuition and user research grow more important as products become AI-native.
• Cutting junior roles is short-sighted: today's juniors build the intuition tomorrow's products depend on.

Enjoyed this one? Find us at conceptocloud.com and on LinkedIn. Subscribe to Engineering Evolved so you don't miss the next episode.

Private equity firms are facing an unprecedented challenge with a backlog of 13,000 companies. The biggest issue for 2025 isn't raising capital or sourcing deals—it's successfully returning capital to investors after buying at market peaks.

Show Notes

Episode Overview

A concise analysis of the private equity industry's current crisis: managing a backlog of 13,000 companies while struggling to return capital to investors.

Key Topics Covered

The 13,000-Company Backlog

• Unprecedented number of portfolio companies awaiting exits
• Industry-wide challenge affecting firms of all sizes
• Redefining what success means in private equity

The Capital Return Challenge

• Why returning capital has become the #1 priority for 2025-2026
• Shift from traditional metrics of success (fundraising and deal flow)
• Impact on limited partners and fund performance

Market Timing Issues

• Consequences of buying at market peaks
• The "top of the bubble" problem
• Current valuation challenges and exit environment

Key Takeaways

1. The private equity industry faces a structural challenge with 13,000 companies in the exit pipeline
2. Capital return has superseded fundraising and deal sourcing as the primary challenge
3. Firms that bought at peak valuations are particularly vulnerable
4. The traditional definition of private equity success is being rewritten

Relevant for:

• Private equity professionals
• Limited partners and institutional investors
• M&A advisors and investment bankers
• CFOs and business owners considering exits
• Financial market analysts

Chapters

• 0:00 - Introduction: The Private Equity Challenge
• 0:11 - The 13,000-Company Backlog Crisis
• 0:19 - Capital Return: The New Priority
• 0:28 - The Peak Valuation Problem

Tom Barber challenges the AI hype cycle, arguing that users care about outcomes, not architecture. Learn why slapping an 'AI-powered' label on everything is the wrong approach, and discover how to thoughtfully integrate LLMs into products without falling into common pitfalls like dependency on unstable APIs or unnecessary chatbot interfaces.

Show Notes

Episode Overview

Tom Barber returns with a critical examination of AI integration in modern software development, challenging teams to focus on user outcomes rather than jumping on the AI hype train.

Key Topics Covered

The AI Marketing Problem

• Why 'AI-powered' labels are often meaningless marketing
• The difference between machine learning (which has existed for decades) and modern LLMs
• Examples of invisible AI: spam filtering, fraud detection, map rerouting
• Users grade products on consistency, not on the impressiveness of the underlying model

Engineering Considerations for LLM Integration

• Choosing the right model for your specific use case (Opus, Sonnet, GPT-4, etc.)
• Tradeoffs between cost, speed, and inference quality
• Building evaluation systems and fallback paths
• Managing latency budgets and graceful degradation
• Handling API outages from providers like Anthropic and OpenAI
• The risks of depending on frontier models that can be deprecated

Trust and Transparency

• AI as a potential trust liability
• Managing user expectations around hallucinations
• The importance of data provenance and quality (garbage in, garbage out)
• When and how to disclose AI usage to users
• The ethical obligation to be transparent when AI makes consequential decisions

Product Strategy

• Why you can't charge an 'AI tax' on top of existing pricing
• Pricing based on outcomes, not on the technology stack
• How to use LLMs to deliver genuine efficiency gains
• Reducing user overhead and friction through thoughtful AI integration

Beyond Chatbots

• Why chatbots may be the most inefficient way to interact with LLMs
• The challenge: How to integrate LLMs without forcing users to type everything
• Asking 'What's now instant that wasn't?' instead of 'How do we add AI?'
• Innovation opportunities for those who can solve the chatbot problem

Key Takeaways

1. Users care about reliable outcomes, not whether you're using AI
2. Engineer for model availability issues and API outages from day one
3. Select and tune models specifically for your use case rather than defaulting to frontier models
4. Be transparent about AI usage, especially for consequential decisions
5. Focus on delivering value through AI rather than adding an 'AI-powered' label for marketing
6. The future belongs to products that leverage LLMs without relying on chatbot interfaces

Resources Mentioned

• Various LLM providers: Anthropic (Claude/Opus/Sonnet), OpenAI (ChatGPT-4)
• Example of model deprecation: Fable model being pulled

Connect

Engineering Evolved is hosted by Tom Barber. If you found this episode valuable, please leave a rating and review to help other leaders discover the show.

Chapters

• 0:00 - Introduction: Users Don't Care If It's AI
• 1:01 - Machine Learning Has Always Been Here
• 2:19 - The AI Marketing Problem: Selling Architecture vs Outcomes
• 5:16 - Engineering Realities: Models, Consistency, and Reliability
• 10:11 - The Cost of the AI Label: Trust and Pricing
• 14:39 - When Users Do Care: Transparency and Consequential Decisions
• 17:15 - Beyond Chatbots: The Future of LLM Integration