The Tech Trek

Dmitri Sedov, Chief Data and Analytics Officer at Allvue, joins to explore why transformation isn’t a destination but an ongoing journey. He shares how financial services are navigating the current wave of AI, what it means to balance short-term expectations with long-term strategy, and why open, modular ecosystems are critical to staying competitive. This conversation goes beyond buzzwords to uncover how leaders can embrace change without getting lost in it.

Key Takeaways

• Transformation isn’t about a single endpoint—it’s about staying curious, iterative, and open to multiple paths.

• AI is accelerating faster than previous technology cycles, but its value lies in solving real customer problems, not chasing hype.

• Companies that lagged in past waves of innovation may now have an opportunity to leapfrog forward.

• The smartest moves aren’t “rip and replace” but incremental improvements that build on existing engines.

• Open, interoperable, and modular approaches reduce risk and keep technology flexible for the future.

Timestamped Highlights

01:26 – Why digital transformation is less about checking boxes and more about a mindset shift

09:47 – The balancing act of managing short-term ROI while exploring long-term AI potential

15:25 – A 24-hour hackathon prototype that changed how Dmitri thinks about multi-agent AI

22:34 – Why interoperability and modularity matter more than big monolithic solutions

27:43 – How to plan roadmaps when technology outpaces predictability

31:43 – The future of “human in the loop” and what workforce transformation might look like

A line that stuck

“Transformation is never about reaching one destination. The more you fixate on a single path, the more likely you’ll be left behind.”

Call to Action

If this episode gave you new perspective on AI and transformation, share it with a colleague who’s wrestling with the same challenges. And don’t forget to follow the show so you’re ready for the next conversation on building smarter, more resilient companies.

What if the key to real work-life balance isn’t about escaping work, but transforming it into your outlet for curiosity and passion? In this episode, Hassaan Raza, co-founder and CEO of Tavus, explores how technology is removing friction in our daily lives and why teaching machines to be more human could unlock new levels of creativity, accessibility, and balance. From redefining what “work you love” actually means to the role AI will play in democratizing opportunity, this conversation challenges assumptions and paints a picture of a future where tech becomes a true partner to people.

Key Takeaways

Work-life balance may not come from hobbies outside work, but from making work itself a fulfilling outlet.

Technology’s real value lies in removing friction and making tools more accessible to everyone, not just the highly technical.

AI and human-computing advances could act as equalizers, giving people with different learning styles or backgrounds the same opportunities.

Hollywood dystopias aside, machines can be designed to replace bad tools, not people—and that creates empowerment, not fear.

The future of tech is less about replacement and more about enhancing human creativity, productivity, and quality of life.

Timestamped Highlights

00:33 — Hassaan explains Tavus’ mission to teach machines how to be human.

04:05 — Why some engineers care more about solving puzzles than the outcomes.

07:08 — How passion-driven work blurs the line between career and hobby.

12:12 — The obsession with removing friction and making machines easier to use.

16:37 — Addressing fears about AI taking jobs and reframing the conversation.

20:57 — How AI can open doors for non-traditional learners and democratize education.

A Line That Stands Out

“If you find work that you really love and are passionate about, that can be your outlet—you don’t need balance because the work itself becomes your balance.”

Call to Action

If this episode gave you a fresh perspective on work, technology, and balance, share it with someone who’s wrestling with the same questions. And don’t forget to follow the show so you never miss the next conversation.

Dane Atkinson, CEO and founder of Odeko, joins the show to unpack the reality of evolving as a founder. He shares why the first idea you start with rarely survives, how to know when it’s time to pivot, and why anchoring on a mission instead of a product keeps you in the game. This conversation dives into frameworks for making hard calls, the messy middle of startup life, and what it really takes to endure as a multi-time founder.

Key Takeaways

• Your first idea probably won’t be the one that works—focus on the customer and the mission, not the concept.

• Pivoting is brutal but necessary; small experiments can create the proof you need to shift direction.

• Founders who learn from failure are more likely to succeed in their second or third ventures.

• Having a North Star rooted in mission makes the day-to-day grind and tough decisions bearable.

• The best outcomes come when investors give founders space to experiment and even fail.

Timestamped Highlights

00:43 – Why Odeko’s mission is to help small coffee shops compete with giants

01:44 – The flawed brilliance of Odeko’s first AI-driven product and the hard pivot that followed

05:28 – The painful trap of chasing product-market fit and the danger of sticking too long

10:24 – Building proof for a pivot and the difference between charisma-driven sales and true demand

14:04 – Why most successful founders are “multi-run players” and what VCs often miss about failure

17:02 – How staying mission-driven keeps founders motivated through setbacks

A line worth remembering

“You can change the product, you can change the delivery, but if you have a North Star that matters, you’ll always know how to steer the company back on track.”

Founder Tip

Test new directions quietly alongside your current model. Early prototypes not only prove viability but also help you win over skeptical teammates, boards, and investors.

Call to Action

If this episode gave you something to think about, share it with a fellow founder or operator who’s in the middle of their own evolution. And don’t forget to follow The Tech Trek so you never miss the next conversation on scaling, leadership, and building companies that last.

What does a day in the life of a startup CTO really look like? Mo El Mahallawy, CTO and co-founder of Shepherd, shares the unfiltered journey from being engineer number one to leading a 50-person Series A company. He opens up about the hardest phases of building, what shifts as your company scales, and how he manages energy, priorities, and mental health along the way. This episode is a practical playbook for any founder, CTO, or tech leader navigating growth.

Key Takeaways

• The early days as a startup CTO are often the hardest—you're coding, recruiting, managing, and wearing every hat at once.

• Growth means trading code for vision: shifting from building features to setting direction and enabling your team.

• Time management is only half the battle—energy management and knowing when you do your best work is equally critical.

• Planning is a mental health strategy: when you control your roadmap, you avoid the burnout of constant reaction mode.

• Taking big swings matters more than just paying down tech debt—bets move the business forward.

Timestamped Highlights

02:41 – The origins of Shepherd and why Mo left Airbnb to build in insurance tech.

07:28 – What makes the early-stage CTO role one of the toughest in startups.

10:44 – Mo’s brutally honest description of those days: “like chewing glass.”

16:45 – How the CTO role evolves post-Series A and the challenges of stepping out of code.

20:44 – Why energy balance beats pure time management.

26:42 – Mo’s take on mental health and how planning became his best defense against fatigue.

A line worth remembering

“Your life is going to suck, and then it’s going to be great—but your job is to make this work.”

Pro Tips

• Use your strongest energy hours for high-leverage work, not busywork.

• Build your network early at iconic companies—you’ll rely on those relationships later.

• Don’t shy away from big bets; they create momentum that tech debt never will.

Call to Action

If you found Mo’s story valuable, share this episode with someone thinking about becoming a founder or CTO. And don’t forget to follow the show on your favorite podcast app so you never miss the next set of scaling playbooks.

What does it take to build a startup in a space where the ground shifts every 90 days? Rahul Sonwalkar, founder and CEO of Julius AI, joins the show to share how he navigates product market fit, resource constraints, and constant model evolution while scaling an AI-first company. Instead of relying on long-term roadmaps, Rahul runs Julius with rapid feedback loops, deep user focus, and a mindset that every team member contributes to AI development. This conversation is a playbook for founders and operators facing uncertainty and speed in equal measure.

Key Takeaways

• Why Rahul believes rigid long-term roadmaps can hold founders back in AI and fast-moving markets

• How limited resources force sharper prioritization, and how to decide what makes the cut

• The two-way product market fit Julius found by serving both data teams and business stakeholders

• Why direct user conversations and dogfooding are non-negotiable for early-stage companies

• A glimpse into the future of BI as AI agents that monitor and surface key metrics automatically

Timestamped Highlights

00:43 – Julius AI’s origin story and how Rahul validated product market fit

05:08 – The surprising way both data teams and business users bring Julius into organizations

09:24 – Why Rahul avoids long-term roadmaps and favors month-to-month iteration

13:55 – The role of feedback loops and customer support in shaping product direction

17:27 – How to prioritize when you only have resources for two out of ten critical features

19:39 – Rahul’s vision for how BI and analytics will evolve with AI agents

A Standout Line

“Overnight, things you thought were impossible for the next 12 months can suddenly become possible. That’s why you can’t have a rigid roadmap when building with AI.”

Founder’s Lesson

Use your own product daily. When the entire team feels the same friction as your users, the right priorities become obvious.

Call to Action

If you’re a founder, operator, or investor looking for practical lessons on building in uncertain terrain, follow the show for more conversations like this. And if Rahul’s approach resonated, connect with him on LinkedIn or explore Julius AI.

Matt Lynch, CTO and co-founder of Sage, joins the show to share how his team is using modern technology to transform elderly care. From rethinking outdated nurse call systems to capturing real-time data that improves both patient safety and caregiver retention, Matt breaks down the technical and human challenges of modernizing this critical industry. This is a story of mission-driven innovation, scale, and the surprising role data plays in quality of life.

Key Takeaways

• The aging population is outpacing caregiver availability, creating an urgent need for tech-driven solutions.

• Most facilities still rely on fragmented or low-quality data, making it difficult to improve care.

• Real-time documentation and streamlined workflows give caregivers credit for their work and reduce burnout.

• New technology can deliver results at a fraction of the cost of outdated systems.

• Adoption succeeds when tech is simple, resident-driven, and seamlessly fits into caregiver routines.

Timestamped Highlights

00:34 — Why Sage was founded: the gaps in elderly care tech that inspired three co-founders

04:36 — The looming scale problem: a 40% surge in the aging population with flat caregiver growth

06:56 — How poor data practices erode care quality—and what synchronous documentation changes

12:05 — Why old systems cost 10x more and how modern tools flip the economics

18:06 — Giving caregivers proof of their work and reducing turnover with better data

21:54 — Lessons in building: why focusing on workflow software beat hardware reinvention

A line worth remembering

“Caregivers finally get credit for the work they’re doing—and that completely changes how families and facilities work together.”

Pro Tip

Start simple. In elderly care tech, the real breakthrough wasn’t reinventing hardware but streamlining the caregiver workflow to make adoption natural.

Stay Connected

If this episode resonated, share it with a friend in tech or healthcare. Don’t forget to follow the show on Apple Podcasts and Spotify so you never miss new conversations with the builders shaping our future.

What does it really mean to build in stealth—and when does it help or hurt? In this episode, Amir sits down with Yoni Michael, co-founder of Typedef, an AI infrastructure startup that recently came out of stealth. Yoni shares why his team chose to stay under the radar early on, how they balanced secrecy with customer discovery, and the lessons they learned about finding product-market fit in a noisy AI landscape. If you’re a founder or tech leader navigating early-stage strategy, this conversation offers practical insights you can apply right away.

Key Takeaways

• Stealth mode isn’t all or nothing—there’s a spectrum between total secrecy and open visibility

• Execution and speed of iteration matter more than protecting “the idea”

• Customer discovery should start before you even write code

• Messaging is never final—test, refine, and keep adjusting as you learn from design partners

• Investors expect shifts at the seed stage, but keeping them in the loop builds trust

Timestamped Highlights

00:38 — Why Typedef chose to launch in stealth and what they’re building in AI infrastructure

04:31 — The double-edged sword of operating in a crowded AI market

09:22 — How Yoni approaches customer discovery without giving away too much

13:55 — Shaping messaging and narrative before coming out of stealth

19:49 — Managing investor expectations when your product vision evolves

25:21 — How to connect with Yoni for advice and community in AI infra

A line that stuck with us

“Your competitive edge isn’t the idea—it’s the ability to execute and course correct fast enough to hit your runway.”

Resources mentioned

• Typedef: typedef.ai

• FENEC open-source framework: [GitHub link from Typedef site]

• Yoni Michael on LinkedIn: linkedin.com/in/yonimichael

Pro Tips

Yoni advises founders to test messaging as early as possible—whether through decks, demo sandboxes, or LinkedIn posts. The feedback loop is as valuable as product feedback.

Stay connected

If this episode gave you something to think about, share it with a founder or tech leader who’d benefit. And don’t forget to follow The Tech Trek on Apple Podcasts, Spotify, or YouTube so you never miss future conversations.

Simone Kalmakis, VP of Engineering at Viam, joins the show to share what it takes to lead teams that bridge hardware and software in the robotics space. With a background in software and machine learning, she now manages engineers with expertise ranging from code to mechanical systems. Simone talks about how she fosters collaboration across disciplines, adapts her leadership style, and keeps her teams aligned while building cutting-edge robotics solutions. If you want to understand what it’s really like to lead at the intersection of AI, software, and hardware, this conversation is for you.

Key Takeaways

• Why managing hardware and software together requires humility, curiosity, and alignment

• How integrating testing across both domains changes the development process

• The role of “20% projects” in helping engineers learn and innovate faster

• Why passion and purpose matter more than hours worked when it comes to work-life balance

• Why strong engineers without PhDs can still excel in robotics and AI

Timestamped Highlights

01:05 — How Viam bridges the gap between physical devices and AI

04:04 — The challenges of testing hardware versus software, and why small changes have big consequences

10:57 — How hobbyist projects and 20% time accelerated Simone’s growth as a leader in robotics

14:09 — What one-on-ones look like when leading both software and mechanical engineers

16:14 — Attracting and retaining top robotics and AI talent in a competitive market

22:54 — How passion and vision drive sustainable work-life balance in high-pressure environments

A Line That Stuck With Us

“Even the slightest change in hardware can have huge ramifications, and it’s humbling to lead people whose expertise far outweighs your own.”

Call to Action

If you enjoyed this episode, share it with a colleague who’s curious about robotics or leadership at the hardware-software frontier. And don’t forget to follow the show so you never miss insights from top tech leaders.

What does it really mean to be data-driven? Mark Gergess, VP of Data and BI at DoubleVerify, joins the show to unpack how data teams can go beyond dashboards to drive meaningful business action. From building an internal consulting lens to evaluating the latest AI tools, Mark shares how his team translates complex data flows into measurable revenue impact. If you’ve ever wrestled with the gap between insights and outcomes, this conversation will hit home.

Key Takeaways

• Being data-driven is about driving action, not just reporting numbers

• Stakeholders don’t care about your data problems—they care about business outcomes

• The biggest challenge with AI adoption isn’t the model, it’s the use cases

• Efficiency gains from AI should shift focus from ETL tasks to solving real business problems

• Data culture health is measured by how naturally teams rely on data day-to-day

Timestamped Highlights

01:17 How DoubleVerify helps advertisers build safer, more effective digital campaigns

04:55 Why the definition of “data-driven” still varies and why it matters

09:25 Measuring whether data efforts are moving the needle on revenue

13:15 How to separate hype from value when evaluating AI and GenAI tools

17:10 Lessons from the data science boom and why companies must go “all in” with AI

25:31 Can AI act as your junior analyst? Where efficiency gains really show up

27:01 How freeing up time changes the structure of data teams and boosts business impact

A thought worth holding onto

“It’s not about dashboards. It’s not about reporting. It’s about doing something with the information.”

Pro Tips

Mark recommends treating AI as a “junior analyst”—let it handle quick, lower-priority questions so your team can focus on bigger business challenges.

Call to Action

Enjoyed the conversation? Share this episode with a colleague who talks about being “data-driven.” Subscribe on your favorite podcast platform and connect with me on LinkedIn for more insights from leaders shaping the future of data and technology.

Teddy Solomon, co-founder and CEO of Fizz, joins the show to share his journey from Stanford dropout to leading one of the fastest-growing Gen Z social platforms. In this conversation, Teddy breaks down what it’s like to learn leadership on the fly, the misconceptions around Gen Z founders, and how he’s scaling a product used by 95% of students at some of the top universities in the country. His story is a rare look at what it takes to build, adapt, and lead with composure at a young age.

Key Takeaways

• How Teddy turned a pandemic side project into the leading college social platform in the U.S.

• Why staying “high with the lows and low with the highs” is his guiding principle for leadership and culture

• The role of mentorship and humility in navigating early CEO decisions

• How to earn respect and lead teams older and more experienced than you

• Why real-world interaction still matters in an era dominated by digital connections

Timestamped Highlights

[01:00] Teddy explains how Fizz became the go-to social app for U.S. college campuses

[02:45] The bold decision to drop out of Stanford after meeting a key mentor and investor

[09:15] Why bringing in an external CEO was the best move for the company’s survival

[12:30] Lessons on leadership: performing your best in the worst moments

[15:59] The reality of leading teams as a young founder

[18:45] Challenging Gen Z stereotypes through action and connection

Memorable Line

“You need to stay high with the lows and low with the highs. If you can perform at your best when things are at their worst, you give your company the best chance to survive.”

Pro Tips

Focus on results and culture. Age, background, and pedigree matter far less than the ability to create a strong team environment and deliver outcomes.

Call to Action

If you enjoyed this episode, share it with someone who’s passionate about leadership and startups. Follow the show for more conversations with founders shaping the future of tech, and connect with us on LinkedIn for behind-the-scenes insights.