
Sign up to save your podcasts
Or


In this episode of Terminal Value, I’m joined by Shaju Puthussery and Deepak Ramaswamy, co-founders of LightSpun, to discuss why parts of dental insurance administration still rely on fax, paper, and legacy core systems, and what it takes to rebuild the back office for AI.
We cover LightSpun’s approach to eligibility, credentialing, claims, and payments; how deterministic rules and AI agents work together; and where customers retain control and licensed clinicians make the clinical decisions.
Shaju and Deepak also explain their credentialing entry point, enterprise adoption and deployment, model selection and data organization, and why they argue that AI-native architecture changes the work of migrating off legacy systems.
The conversation closes with LightSpun’s cost thesis, its plans to expand beyond dental into other benefit categories, and the founders’ views on how quickly regulated industries can change.
LightSpun: https://www.lightspun.com/
Shaju Puthussery: https://www.linkedin.com/in/shajuputhussery/
Deepak Ramaswamy: https://www.linkedin.com/in/deepak-ramaswamy-0124032/
Terminal Value: https://www.youtube.com/@Terminal-Value-Podcast
Chapters
0:00 Why legacy insurance systems are hard to adapt for AI
0:51 Introducing Shaju Puthussery, Deepak Ramaswamy, and Lightspun
1:44 Why dental insurance still relies on fax and paper
7:13 Rebuilding the dental insurance back office
11:20 Automating claims while preserving clinical judgment
17:51 How better models and organized data expand automation
22:02 Winning customers through credentialing and industry partnerships
26:35 Adapting deployment to each customer's systems
28:16 Competing with homegrown systems and legacy vendors
30:41 How AI helps migrate data and configure business rules
33:56 Expanding from dental into vision, pharmacy, and other benefits
38:22 How quickly regulated industries can modernize
40:50 Closing takeaways
Disclaimer: For information and education only; not investment advice or a recommendation to buy or sell securities. Views are the speakers’ own.
In this episode of Terminal Value, I’m joined by Karl Alomar, Managing Partner at M13 and former Chief Operating Officer of DigitalOcean, to discuss where value may accrue as AI models become cheaper, why proprietary context and data matter, and what makes an application difficult to replace.
We cover DigitalOcean’s developer-first strategy and just-in-time capacity planning, the GPU shortages that gave neoclouds their opening, inference efficiency, model pricing, real-time data infrastructure, and AI products embedded in core business operations. Karl also shares his view on AI-enabled microbusinesses, the value of operating experience in venture capital, the tension between company-building and secondary liquidity, and the identity and security layers AI agents may need.
Terminal Value: https://www.youtube.com/@Terminal-Value-Podcast
Chapters
0:00 Why the real AI moat sits outside the model
0:35 Introduction to Karl Alomar and M13
1:30 Why DigitalOcean succeeded against larger cloud providers
3:50 GPU scarcity, neo-cloud economics, and capacity planning
13:50 How software can dramatically improve GPU efficiency
20:55 Open-source models, pricing pressure, and application-layer moats
31:45 Real-time data infrastructure and AI embedded in business workflows
38:40 How AI could reshape employment and create millions of micro-businesses
42:40 Why operator experience matters as venture funds get bigger
48:15 Private AI valuations, secondary markets, and the challenges of going public
57:40 How experience shapes investors and the value of ambitious young founders
1:01:05 Why AI agents will need identity, verification, and security
1:02:50 Closing takeaways
Disclaimer: For information and education only; not investment advice or a recommendation to buy or sell securities. Views are the speakers’ own.
Work happens across Slack, Notion, Linear, GitHub, meetings, and the desktop. The problem is not a lack of tools. It is the context lost between them.
In this episode, Nik Singh sits down with Parris Khachi to explore Highlight's approach to proactive workplace AI. Parris explains why the next AI interface may not be a chat box and how Highlight uses screen, meeting, and integration context to make assistance more proactive.
The conversation also covers privacy and user control, individual versus team context, model economics, product-led adoption, competition with incumbents, and why teams should automate coordination without outsourcing judgment.
Highlight AI: https://highlightai.com/
Parris Khachi: https://www.linkedin.com/in/parris-k-7928b011/
Terminal Value: https://www.youtube.com/@Terminal-Value-Podcast
Chapters
00:00 Cold open
00:56 Introduction
01:31 Where context gets lost across work tools
05:03 Highlight's mission and the coordination tax
06:04 The AI intern that follows the work
13:01 Screen context, privacy, and user control
18:18 Choosing models across capability, speed, and cost
21:43 Ideal users, deployment, and adoption
27:28 How AI could change team structure
29:32 Competition and the interface after chat
33:51 Highlight's long-term vision
36:10 Closing takeaways
Disclaimer: For information and education only; not investment advice or a recommendation to buy or sell securities. Views are the speakers' own.
In this episode of Terminal Value, I'm joined by Cy Khormaee, founder and CEO of Aegis AI, an AI-native email security company that recently raised a $36 million Series A led by Battery Ventures, to discuss why email security is becoming an AI-versus-AI problem, how personalized threats are bypassing legacy rules, and what changes when AI begins handling the intelligence layer of cybersecurity.
We cover how email security evolved from reputation-based appliances to machine learning and rules, why LLM-based reasoning can investigate each message in context, how Aegis uses specialized multi-agent models instead of frontier APIs, why proprietary email data creates a compounding advantage, how automated incident response can shrink exposure from hours to milliseconds, what security buyers need to trust autonomous systems, why AI-native architecture can challenge incumbents, how the category could expand from protecting the corporate inbox to protecting the person, and why Cy hires for slope in an uncertain future.
0:00 Hook: Why email security is becoming AI versus AI
0:40 Intro to Cy Khormaee and Aegis AI
1:45 How email security evolved from reputation and rules to AI
4:55 How AI is making phishing attacks more personalized and harder to detect
9:35 How Aegis uses multi-agent reasoning to stop email threats
12:50 Why rule-based security breaks as threats change faster
15:45 How proprietary email data creates a compounding AI advantage
18:00 How automated response changes accountability and earns trust
25:10 Why Aegis builds specialized models instead of relying on frontier models
31:45 How AI-native architecture can overcome incumbent data advantages
39:25 From securing email to protecting the person
41:15 Why Cy hires for slope in an uncertain AI future
42:50 Closing takeaways
In this episode of Terminal Value, I'm joined by Dylan Robbins, founder and CEO of Lucra, which recently raised $20 million led by ARK Invest Venture Fund to build competitive loyalty infrastructure, to discuss why traditional points programs often struggle to drive adoption, how games and immediate rewards can make loyalty feel worth using, and why the real moat sits in the payments, identity, fraud, and compliance stack underneath the experience.
We cover Lucra's shift from a consumer app to an embedded B2B platform, how the company sells to venues and proves payback, why customers that try to build the stack themselves often come back, how Crown brings mobile games into the days between venue visits, how Bracketology extends the model into reality TV, what loyalty could look like when rewards are narrower and immediate, and why Dylan thinks the AI funding boom may be crowding out other kinds of innovation.
0:00 Hook: Why building loyalty in-house costs more than brands expect
0:45 Intro to Dylan Robbins and Lucra
1:43 Why traditional loyalty programs struggle to drive adoption
4:17 How Lucra replaces points with immediate rewards
6:23 From customer goals to integration, payback, and ROI
11:20 Market size, pricing, and the path to self-serve integration
15:22 Lucra’s three growth pillars: IRL, mobile games, and reality TV
17:26 Why brands eventually buy instead of build
21:12 How Crown connects physical venues to mobile games
25:27 How Bracketology expands Lucra into reality TV
30:09 What loyalty could look like in five years
31:54 Why the AI funding boom may be hurting startup innovation
34:05 Closing takeaways
In this episode of Terminal Value, I'm joined by Viral Shah, co-founder and CEO of JuliaHub, the company behind Dyad — a physics compiler paired with AI agents for engineering and simulation — to discuss why physical product engineering has been so hard to automate, what an AI agent actually needs to model the real world, and whether hardware development can start to look much more like software.
We cover the "engineering V" and why every step runs in its own siloed tool, why the real TAM is the trillions spent designing products rather than the billions spent on simulation software, how Dyad works as a compiler that fuses disparate physics into one simulation, why calibration — not modeling — is the hard part, how Instron cut crash-test-rig calibration from over a day to under a minute, why trusted reusable model libraries make hardware feel like software (text, Git, CI/CD, agents), how agent pricing shifts from per-seat licenses to token consumption, why the harness can matter more than the frontier model, why engineering shows high expected but low observed AI impact today, and why Viral believes hardware engineering is going to become like writing software.
0:00 Hook: Why hardware engineering could become like software
0:35 Intro to Viral Shah, JuliaHub, and Dyad
1:25 How engineers move from product requirements to validated designs
4:41 Why engineering software is fragmented across specialized tools
12:16 Where Dyad fits and why JuliaHub built a physics compiler
19:17 How real-world data calibrates simulation models
23:21 Building trusted models that can be reused like software libraries
26:17 Where Dyad is being used across aerospace, semiconductors, and infrastructure
29:10 How agentic engineering changes seat-based software pricing
31:37 Why the agent harness can matter more than the frontier model
35:44 Why AI adoption in engineering still lags software
38:13 Why Viral Shah believes hardware development will become more like software
41:04 Closing takeaways on Dyad and the shift toward computational engineering
Disclaimer: For information and education only; not investment advice or a recommendation to buy or sell securities. Views are the speakers’ own.
In this episode of Terminal Value, I'm joined by David Sosna, founder and CEO of Sympera AI, which raised a $10M seed round to bring agentic AI into relationship banking, to discuss why a copilot is the wrong model for enterprise AI, whether AI replaces relationship bankers or expands the market they serve, and why the real win is embedding AI into the banker's process instead of bolting on a chatbot.
We cover why relationship banking is a large, human, relationship-driven market, why more query and data tools never fixed the banker's day, how Sympera narrows a huge book down to a few fully-researched opportunities and tells the banker what to do next, how it blends public, purchased, and internal bank data without ever exposing what the bank holds, why betting on a single AI model is risky and how a multi-model stack with tuned tools answers it, why running everything through the most expensive frontier model doesn't scale, why banks that try to boil the ocean get nothing done, and why vertical AI's real edge is embedded judgment rather than a better chat box.
0:00 Hook: Why AI copilots fail without process redesign
0:45 Intro to David Sosna and Sympera AI
1:29 How relationship bankers grow and protect client relationships
4:04 Why CRM and data tools still depend on banker judgment
9:28 How Sympera narrows a market to five actionable opportunities
12:43 From bank priorities to ranked leads and next-best actions
16:39 Data security, lead accuracy, and continuous model validation
19:47 Feedback loops, inference cost, and scanning at scale
24:46 Why Sympera uses multiple models and keeps outreach human-reviewed
29:22 Bank targeting, pricing, and the work of changing user behavior
34:53 Salesforce, vertical AI, and the future banking agent stack
41:38 AI-native teams, banker productivity, and why copilots fall short
49:05 Closing takeaways
Disclaimer: For information and education only; not investment advice or a recommendation to buy or sell securities. Views are the speakers’ own.
In this episode of Terminal Value, I'm joined by John Taylor Garner, founder and CEO of Odynn, an AI-native travel and loyalty infrastructure company that has raised over $9M in seed funding from Bonfire Ventures and Fiat Ventures, to discuss why points and miles behave less like a rewards program and more like a private economy, why the real product in travel is the decision layer rather than the booking, and how Odynn turns a multimillion-dollar embedded-travel build into something a late-seed startup can afford.
We cover how the fragmented travel rewards stack works today, why Expedia and Booking quietly control most of embedded travel, why points and miles only ever depreciate, how Odynn acts as a "Shopify for travel rewards" that compares dollars, points, and miles in one place, how the company prices and earns alongside its bank customers, what it takes to sell to banks as a young startup, and how an agentic AI concierge could turn travel from something you search into something you delegate.
0:00 Hook: Airlines don't make money flying planes
0:45 Intro to John Taylor Garner and Odynn
1:35 How the travel rewards stack works today
3:25 Banks, OTAs, and the Expedia/Booking duopoly
5:55 Where it breaks down for the consumer
8:10 Why points and miles only lose value
10:35 What Odynn is and how the product works
16:15 Pricing, commissions, and supply economics
20:30 Customers, competition, and the $60K disruption
25:00 Winning big banks and proving longevity
30:00 The AI concierge and the future of loyalty
36:00 Building AI-native: agents, engineers, and OpEx
41:55 Closing takeaways
Disclaimer: For information and education only; not investment advice or a recommendation to buy or sell securities. Views are the speakers’ own.
In this episode of Terminal Value, I'm joined by Curt Aubley, co-founder and CEO of Sevii, and Stephen Collins, co-founder and CTO. Sevii is building autonomous defense and remediation for security operations, backed by Overline and BMW i Ventures. We discuss why the SOC was built for a slower world, what separates autonomous remediation from automated playbooks, and whether companies are culturally ready to let AI take real security actions.
We cover the legacy SOC workflow from alert to remediation and why it hands attackers a 15-minute head start, how Sevii's agentic "cyber warriors" go from detection to reported remediation in 2 to 15 minutes across the tools you already own, why per-asset licensing keeps security budgets predictable, Sevii's rule of only using AI where it's necessary, and why the hardest competition may be cultural trust rather than another vendor.
0:00 Hook: Why attackers get a 15-minute head start
0:40 Intro to Curt Aubley, Stephen Collins, and Sevii
1:26 How legacy SOC workflows reach remediation
4:35 Where speed, scale, and economics break the SOC
7:00 Why existing security tools still leave a bottleneck
9:29 What Sevii's autonomous remediation platform does
15:02 How Sevii moves from detection to remediation
19:31 How cyber warriors change SOC staffing
24:00 Customer control, MDR limits, and AI cost strategy
31:20 Sevii's ICP, buyer, and forward-deployed model
37:33 Asset-based pricing and predictable budgets
40:53 Competition, culture, and the future of autonomous defense
48:35 Closing takeaways
Disclaimer: For information and education only; not investment advice or a recommendation to buy or sell securities. Views are the speakers’ own.
In this episode of Terminal Value, I'm joined by Tushar Makija, co-founder and CEO of Team Ohana, to discuss why headcount planning breaks the moment the CFO's model is locked, how AI agents are becoming the orchestration layer between HR and Finance, and why the real system of record should capture decisions, not just transactions.
We cover why a late hire is deferred capital, how a sales leader can plan ten hires from a single Slack prompt, why HR and Finance never demand better tools, human capital versus "agentic capital," and Tushar's "accelerate or die" case for moving now.
Disclaimer: For information and education only; not investment advice or a recommendation to buy or sell securities. Views are the speakers’ own.
From the publisher's feed