# Shipping like a startup, regulated like a bank — David Abitbol — session 2026-10-13T19:20:00.000Z → 2026-10-13T19:38:00.000Z _31 transcript lines · 54 slides · source: full recording_ ## Transcript **[00:00:00 · 0]** As we know throughout time, the only constant in this world has been change. What's different in the AI and super intelligence era is that intelligence has become cheap. Regardless of what your unit of intelligence is, whether it be software, analytics, recommendations, reports, AI is creating an abundance in all of it. And, as we know, more answers don't necessarily create more value. So what is scarce? **[00:00:34 · 0]** What is a scarce resource in an abundant world? What what are companies taking advantage of, the leaders at least? It's judgment. What is judgment? Judgment is choosing what matters and how to act when there's no single obvious answer. **[00:00:55 · 0]** It's the creativity that lets you explore what possibilities are worth exploring. It's the taste that tells you what good looks like in your given context. It's the decisions where you can weigh what trade offs are worth making. Judgment is ubiquitous. It exists throughout your organization on a daily basis. **[00:01:17 · 0]** Your c suite... The c suite is asking questions, taking advantage of judgment judgment all the time. So are the software engineers, the AI leaders, the portfolio managers, and the data scientists. So the question that they're focused on is given judgment is a finite asset, a limited capacity, the main question becomes, where is our best judgment worth spending? Like for you to hold on to that thought. **[00:01:50 · 0]** With that, we'll dive into our first pillar, design the right problem. So questions, a question that, the leading organizations are asking all the time is they're obsessed over what is the right problem to solve, right? It's tempting to start with the tool. Today, we're surrounded with better models, harnesses, we're surrounded with all these AI tools, we're surrounded with co pilots, But that isn't the way about, is it? At least that's not the way the leaders are approaching it. **[00:02:24 · 0]** Now let's go back to our hypothetical, two companies. Let's say they're from asset management. Let's say, we're tracking the workflow of two research analysts in these hypothetical organizations. Well, let's say organization b starts with the tool, the copilot. They're asking, where can we add AI to the analyst work? **[00:02:47 · 0]** They're looking at the current existing process to say, earning transcript comes in. I discovered a summarization workflow. Let me pass all of this in through my fire hose. Analysts get earning summaries faster. Win win, right? **[00:03:05 · 0]** Well, not so much, at least not compared to the guys who are leading. And such a workflow of of being a faster horse introduces one key, nuance. That is you are losing context as you have information, I guess, reduction as part of your summarization. So what what is a hypothetical organization, a, doing in this? Well, they're obsessed in the right problem to ask. **[00:03:31 · 0]** They could be asking, why are our material signals reaching us too late? Where do we focus the attention, the AI, the bandwidth, the capability on? Where is our bottleneck? Right? And so it becomes this redesigning of the workflow or at least the reevaluation on what to focus on. **[00:03:50 · 0]** That's what they're doing different. Now this consideration isn't particular to finance. It's across the the realm and function. You know, you're asking the same question on what to focus on, whether it be research capacity, decision speed, quality of your data, latency, reliability. It it it traverses all of that. **[00:04:12 · 0]** And studies show this. They showed leading studies, on average spend their attention on fewer higher quality decisions than the other cohort. So leaders are being more selective in where they focus their attention and what they focus their attention. So it's great. The leading companies, have, an idea of what the right problems are. **[00:04:37 · 0]** Where do they go from there? Well, they also concentrate their investments in the highest value workflows. I get it and I've been, I've been culpable of of something similar where, you know, is AI is almost creating too many opportunities and experimentation can be a good thing. But one needs to separate experimentation and the resulting innovation theater or worse, the productivity theater, which actual value generation. Going back to our two, hypothetical companies, what does this mean? **[00:05:13 · 0]** Company b is looking at how do we launch the most number of AI pilots. And what that ends up looking like is a document summarization bot here, a copilot there, a data retrieval bot here. It's scattered all over the place. Whereas what company is focused on in their investments is potentially use cases like end to end workflows on the research, on the research pipeline. They're focused on aligning engineering, analytics, AI, capacity, and judgment towards a singular focus. **[00:05:51 · 0]** And research has shown that the leaders spend more than 80% of their investments on these high value use cases, transforming key business functions and creating new offerings. Perfect. We we know they're focusing on, finding the right problem and they're also spending deep investment in select workflows. Well, they're also doing something different. They're reimagining how work is done. **[00:06:25 · 0]** I think this is this is key and it goes back to what we discussed previously, which is judgment. They're redefining and reorganizing functions, roles, capacities around this key element of judgment. What does that mean for our hypothetical company? Well, company b might be focused on, as we discussed, I'm getting my earnings transcript and getting my research, my research reports. How do I summarize it? **[00:06:52 · 0]** Maybe I make that faster. Maybe I update my financial models faster. Maybe I send my recommendation to my PM faster. Company a is focused on an entirely different ball game. Hypothetically, they could be asking the analyst to define the investment thesis and they're watching and gathering evidence that could refute that thesis or support it. **[00:07:15 · 0]** They'd be monitoring continuously, they'd be investigating exceptions and they could be deep diving on, you know, select examples that could lead to actual generation of value. So the winners just don't just automate, they redesign. And studies show that the cohort of high performers are 75% more likely or in practice, they actually do redesign their workflow versus 25% of the cohort that doesn't. So where does this leave us with our two hypothetical companies, and designing the right problem? Company b adds AI to existing workflows. **[00:07:57 · 0]** They're completing individual projects and they're moving to the next initiative. Whereas company a is rethinking how work is done. They're redesigning processes around AI and they're obsessing of the the the right problem. That's that's the difference in how organizations puts AI to work. Great. **[00:08:15 · 0]** So they picked the right problem. What what happens next? But we all know in order to actually enable that capacity, you need to build share capabilities across the org. And what's what's a trend that's seemingly emerging for for some time now is that the lines between technology and business are converging. And we're seeing this everywhere. **[00:08:40 · 0]** We're seeing engineers that are moving closer to business outcomes. We're seeing business professionals that are building tools and automations. And as you can see and imagine, you can have forward deployed engineers, you can have forward deployed product managers, but you could also have the inverse which is forward deployed, portfolio managers. And it's really this combination of getting the right people in the room by breaking down organization silos that the leaders are really leaning on. Studies have shown that, more and more of the large enterprises are viewing their technology arm as a business growth function, and that's creating an edge for these for these companies. **[00:09:26 · 0]** So the other sort of vertical inside building shared capabilities that, is probably a personal favorite of mine, it's standardizing the core and personalizing the edges. And this is something that isn't new, to the technology world. We've we've built infrastructure. We've built, you know, shared modules and system design. But what's interesting here today is that's taking a very different shape with the capabilities out there and with the org and the breaking down of silos. **[00:09:56 · 0]** And this could look very different depending on where you sit. You could be building shared course for a, you know, a team that looks very different with analysts and and clients. You could be doing that with, you know, a bunch of analysts or you can do... Be doing that for software engineers or or data scientists. With the data scientists example in particular, that could look very interesting where instead of having one off workflows that that is really common in the industry, you can build a shared core of the datasets, their semantic layers, the trade offs of what worked, what didn't, what, you know, what what some datasets do good of, where are the gaps and and so on and so forth. **[00:10:41 · 0]** Such that each new personnel becomes almost a plug and play module in the system where they can then stand on the shoulder of their predecessors. So what should the next person not have to rebuild? Now what's also interesting is, you know, the enterprise wasn't really built for agents. Agents, they they work across workflows and applications. They they take context. **[00:11:10 · 0]** They, deliberate. They take actions. Whereas, really, the enterprises are focused, previously on, single human or a collection of them interacting with a single application. So the leaders are cognizant that something needs to change and that they need to modernize, their infrastructure. And they're very self aware of where they are in their journey. **[00:11:36 · 0]** And so studies show that, you know, 75... 71% of the orgs are modernizing their core infrastructure to support AI. And, you know, six to 10%, or 23% of them are investing six to 10% of their annual revenue in in core system modernization, to take advantage of this. And a common complaint here or a common challenge here is, you know, data quality and and the ability to reuse data. So great. **[00:12:06 · 0]** We, you know, we we have the right people in the room. We, have, you know, we have the right knowledge there. What does... What do we do next? And then comes the concept of, hey, how do we make this knowledge reusable and accessible? **[00:12:20 · 0]** And that's where we have, you know, depending on on the adoption and where where they are, you can have different types of team agents where you can have expert agents being subject matter expert. You can have workflow agents that work across streams like Slack bots that are creating and opening tickets. They're closing them and what have you. And then you can have really chief of staff kind of agents that keep a tab on what the state of the organization is and what needs to be done. So, you know, going back to our workflows, a, again, is looking at this very differently in terms of shared capabilities. **[00:12:56 · 0]** Now the last pillar here that I wanna draw, your attention to is, well, when does this system really compound? What happens when it's deployed? Well, leading companies, what they're doing is they're actually thinking about it slightly differently. What are failures that can give us an edge that we can learn from? If if there's a success, how do you make it part of the system? **[00:13:22 · 0]** And the way to think about this is, you know, the individual learned but did the system. Is there a way perhaps when they've encountered something where they need to make a correction that that knowledge is fed back into the org either through a new tool or context or or or what have you. And, what's what's key here is preserving the evidence along with those learnings. So, really the last vertical here is AI freeze capacity, where does it go, being cognizant that you need to redesign your system and really focus it on key metrics that you're evaluating, you're monitoring, and you're creating a fry... Flywheel off. **[00:14:08 · 0]** So where does this leave us for companies a and b? You know, company a has their entire work redesigned. They're building capabilities at compound, and they have shared capabilities built. So I wanna leave you off with one final question. Everyone has AI who has the advantage. **[00:14:27 · 0]** Thank you. ## Slides ### 00:00:00–00:00:20 # Build the Competitive Edge 1. Design the right problem 2. Build shared capabilities 3. Compound the advantage ## Everyone Has AI. Who Has the Advantage? Sid Pednekar / Head of Data and AI, Asset Management ### 00:00:20–00:00:40 # Intelligence is becoming cheap AI is creating abundance in software, analysis, recommendations, and outputs. But more answers don't necessarily create more value. ## What is the scarce resource in an ### 00:00:40–00:01:00 # Intelligence is becoming cheap AI is creating abundance in software, analysis, recommendations, and outputs. But more answers don't necessarily create more value. ## What is the scarce resource in an abundant world? 1. Design the right problem 2. Shared capabilities 3. Compound ### Everyone Has AI. Who Has the Advantage? Sid Pednekar / Head of Data and AI, Asset Management ### 00:01:00–00:01:20 # Judgement Choosing what matters and how to act when there is no single obvious answer. ## Creativity Which possibilities are worth exploring? ## Taste What does good look like in this context? ## Decisions Which trade-offs are worth making? ### Everyone Has AI. Who Has the Advantage? Sid Pednekar / Head of Data and AI, Asset Management [Horizontal timeline with three steps: "01 Design the right problem", "02 Shared capabilities", "03 Compound", and indicators for "Company A", "Company B", and "A/B"] ### 00:01:20–00:01:40 # The ubiquity and scarcity of judgement Judgement runs through the work. Human attention is limited. - **CTO** What should we build next? - **Software Engineer** Which approach fits the problem? - **Portfolio Manager** Which risks are worth taking? - **COO** Where should we add capacity? - **AI Leader** Which problems are worth pursuing? - **Data Scientist** Is the evidence good enough to use? ### Where is our best judgement worth spending? 1. Design the right problem 2. Shared capabilities 3. Compound - Company A - Company B - A/B ## Everyone Has AI. Who Has the Advantage? Sid Pednekar / Head of Data and AI, Asset Management ### 00:01:40–00:02:00 # The ubiquity and scarcity of judgement Judgement runs through the work. Human attention is limited. - **CTO**: What should we build next? - **COO**: Where should we add capacity? - **SOFTWARE ENGINEER**: Which approach fits the problem? - **AI LEADER**: Which problems are worth pursuing? - **PORTFOLIO MANAGER**: Which risks are worth taking? - **DATA SCIENTIST**: Is the evidence good enough to use? ## Where is our best judgement worth spending? ### 00:02:00–00:02:20 # Design the **right problem** ### 00:02:20–00:02:40 ## PART I / Target high-value business problems # Leading organizations are obsessed over the right problems to solve ### 00:02:40–00:03:00 ## PART I / Target high-value business problems # Start with right problem. ## COMPANY B / AI TOOLS FIRST > "Where can we add AI to analysts' work?" ### 00:03:00–00:03:20 ### PART I / Target high-value business problems # Start with right problem. ### COMPANY B / AI TOOLS FIRST - "Where can we add AI to analysts' work?" - ### 00:03:20–00:03:40 PART I / Target high-value business problems # Start with right problem. ### COMPANY B / AI TOOLS FIRST > "Where can we add AI to analysts' work?" More ### 00:03:40–00:04:00 # Everyone Has AI. Who Has the Advantage? ## PART I / Target high-value business problems ### Start with right problem. **COMPANY B / AI TOOLS FIRST** > ### 00:04:00–00:04:20 ### Design the right problem ## 2. Solve the Real Bottleneck Find what actually constrains performance, not necessarily what's easiest to automate. ### BUSINESS LENS - Research ### 00:04:20–00:04:40 # Leading companies start with the right problem ## Average number of prioritized AI use cases - **3.5** Leading companies - **6.1** Other companies ## Leaders are more selective about where they focus. BCG, From Potential to Profit—Closing the AI Impact Gap (Jan. 2025) Survey of more than 1,800 executives ### 00:04:40–00:05:00 ### PART I / Target high-value business problems # Leading companies start with the right problem ## Average number of prioritized AI use cases - **3.5** Leading companies ### 00:05:00–00:05:20 # PART I / Concentrate Investment ## They concentrate investment in their highest value workflows ### Agenda - 01 Design the right problem - 02 Shared capabilities - 03 Compound ### 00:05:20–00:05:40 ### PART I / Concentrate Investment # Concentrate AI investment where it matters ## COMPANY B / SCATTERED PILOTS > "How many AI pilots can we launch?" Many pilots. ### 00:05:40–00:06:00 PART I / Concentrate Investment # Concentrate AI investment where it matters ### COMPANY B / SCATTERED PILOTS > "How many AI pilots can we launch?" Many pilots. Limited compounding value. ### COMPANY A / FOCUSED PROGRAM > "Which workflow is worth scaling?" One scaled capability. Reusable advantage. ## Everyone Has AI. Who Has the Advantage? Sid Pednekar / Head of Data and AI, Asset Management [Illustration showing five individuals working independently on laptops, with data flowing to and from them, representing scattered pilots.] [Illustration showing a team of three collaborating around a central data system with multiple screens and ### 00:06:00–00:06:20 # Concentrate Investment ## Leaders concentrate AI investment In their highest value workflows > **>80%** of leaders' AI investment > > Transforming key business functions and creating new offerings ### 00:06:20–00:06:40 ### PART I / Redesign workflows # They are rethinking how work is done ## Everyone Has AI. Who Has the Advantage? Sid Pednekar / Head of Data and AI, Asset Management [Horizontal progress bar with three steps: "01 Design the right problem" (highlighted), "02 Shared capabilities", and "03 Compound". Below the progress bar, a slider compares "Company A" and "Company B".] ### 00:06:40–00:07:00 # Everyone Has AI. Who Has the Advantage? Sid Pednekar / Head of Data and AI, Asset Management ## PART I / Redesign workflows ### Strategy Where should ### 00:07:00–00:07:20 PART I / Redesign workflows # The winners don't just automate. **They redesign.** ## COMPANY B: REACTIVE ### COMPANY B / EVENT-DRIVEN 1. Earnings ### 00:07:20–00:07:40 # The winners don't just automate. **They redesign.** ### PART I / Redesign workflows ## COMPANY B / EVENT-DRIVEN 1. Earnings trigger the work (Analyst ### 00:07:40–00:08:00 ### PART I / Redesign workflows # The winners don't just automate. **They redesign.** Organizations that fundamentally redesigned workflows because of AI - **25%** Other organizations - **NEARLY 75%** AI high performers - **≈3x** as likely to report workflow redesign High performers: defined using reported enterprise financial impact. Source: McKinsey, The State of AI, 2026. --- **Current Section:** 01 Design the right problem ### 00:08:00–00:08:20 ### PART I / The two companies # Design the right problem ## Company B - Adds AI to existing workflows. - Completes individual projects. - Moves to the next ### 00:08:20–00:08:40 ## PART II # Build Shared Capabilities [Progress bar showing three steps: 1. Design the right problem, 2. Shared capabilities (highlighted), 3. Compound. A legend indicates ### 00:08:40–00:09:00 ### PART II / Technology and business # 1. Technology and business are converging How do we move from one successful workflow to an organization that can deliver them repeatedly? ## TECHNOLOGY ### 00:09:00–00:09:20 ## PART II / Technology and business # 1. Technology and business are converging How do we move from one successful workflow to an organization that can deliver them repeatedly? ### TECHNOLOGY - ### 00:09:20–00:09:40 ## PART II / Technology and business # Technology is becoming a business growth function **66%** of large enterprises view their technology organizations as revenue generators rather than service centers Technology leadership is ### 00:09:40–00:10:00 PART II / Standardize the core # 2. Standardize the core ## Personalize the edges Different applications should not require rebuilding the same foundation - **Research assistant** ### 00:10:00–00:10:20 ## PART II / Standardize the core ### 2. Standardize the core Personalize the edges What should the next person not have to rebuild? Everyone Has AI. Who Has ### 00:10:20–00:10:40 PART II / Standardize the core ## 2. Standardize the core ### Personalize the edges What should the next person not have to rebuild? Everyone Has AI. ### 00:10:40–00:11:00 ### PART II / Standardize the core ## 2. Standardize the core Personalize the edges What should the next person not have to rebuild? Everyone Has AI. Who ### 00:11:00–00:11:20 ## PART II / Enterprise systems and data ### 3. Enterprise modernization and the data bottleneck Agents need to retrieve information and act across systems designed around people using individual applications #### THE MODEL CAN CHANGE - A better model should fit the system #### THE ORGANIZATION MAINTAINS - Reliable data access and business context - Permissions and reusable tools - Orchestration and monitoring - Evaluation of what good looks like ### The system around the model matters just as much [A horizontal progress bar or timeline with three steps: "01 Design the right problem", "02 Shared capabilities", "03 Compound". Step 2 is highlighted with a marker labeled "B". Below the bar, there are two points labeled "Company A" and "Company B".] ### 00:11:20–00:11:40 # 3. Enterprise modernization and the data bottleneck ## PART II / Enterprise systems and data Agents need to retrieve information and act across systems designed around people using individual applications ### THE MODEL CAN CHANGE ### 00:11:40–00:12:00 ### PART II / Enterprise systems and data # Agents need modern infrastructure and usable data ## CORE-SYSTEM MODERNIZATION - **71%** of surveyed organizations are modernizing core infrastructure to support ### 00:12:00–00:12:20 # Agents need modern infrastructure and usable data ## CORE-SYSTEM MODERNIZATION - **71%** of surveyed organizations are modernizing core infrastructure to support AI - **23%** of organizations invest **6–10%** of annual revenue in core-system modernization ## CHALLENGES TO AI AUTOMATION - **48%** cite difficulty finding or discovering data (searchability) - **47%** cite difficulty reusing data (reusability) Sources: Deloitte, Tech Trends 2026: The Great Rebuild (71%; 23% investing 6–10%). Deloitte, Tech Trends 2026: Agentic AI Strategy (48%; 47%, from a 2025 Deloitte survey). ### 00:12:20–00:12:40 ## PART II / Shared context and team agents # 4. Shared context and team agents Make relevant knowledge accessible where needed - **01 EXPERT** - Make know- ### 00:12:40–00:13:00 ### PART II / Shared context and team agents # 4. Shared context and team agents ## Make relevant knowledge accessible where needed - **01 EXPERT** - Make know ### 00:13:00–00:13:20 # Build shared capabilities ## PART II / The two companies ### Company B - Business and technology work apart. - Each team rebuilds its own tools. - Data access stays project-specific. - Know- ### 00:13:20–00:13:40 ### PART III # Compound **the advantage** 1. Design the right problem 2. Shared capabilities 3. Compound [Timeline showing three numbered stages: Design the right problem, Shared capabilities, and Compound, with markers for Company A and Company B] ### 00:13:40–00:14:00 ### PART III / Preserve and curate what you learn # The number survives. # The caveat disappears. Each handoff can preserve the conclusion while losing its meaning. ## ANALYST NOTE ### 00:14:00–00:14:20 # Same AI. Different organizations. Now a better model arrives. Both companies benefit. **COMPANY B** - **Productivity improves.** - **Much of the progress stays local.** - Tools layered on. Projects completed. **COMPANY A** - **Work redesigned** - **The organization's capability compounds.** - **Shared capabilities built** The advantage is being built to turn better AI into more value. Everyone Has AI. Who Has the Advantage? Sid Pednekar / Head of Data and AI, Asset Management ### 00:14:20–00:14:40 ### CONCLUSION / One final comparison # Same AI. Different organizations. Now a better model arrives. Both companies benefit. **COMPANY B** - Productivity improves. - Much of ### 00:14:40–00:15:00 # AI Engineer ## NEW YORK PRESENTed BY **Arize** [Stylized art deco cityscape graphic] ### 00:15:00–00:15:20 # AI Engineer Code Summit ### Presented By Google DeepMind ### November 10-12, 2026 SF // CA AIE CODE returns for its second annual code-focused summit. Catch up on the year with the world's top agentic engineers, sharing the workflows and breakthroughs changing how software gets built. ### Learn More At [https://ai.engineer/code](https://ai.engineer/code) ### Supported By G2i [Nighttime cityscape with a bridge and binary code overlay] ### 00:15:20–00:15:40 # AI Engineer Europe 2027 ## ExCeL London ### 02.17 - 02.19 2027 Europe's flagship technical ### 00:15:40–00:16:00 # PRESENTED BY ## Arize View Full Schedule at [ai.engineer/nyc/2026/schedule](ai.engineer/nyc/2026/schedule) [Arize logo, a stylized pink 'A' shape] ### 00:16:00–00:16:20 # AI Engineer Europe 2027 ## AI Engineer Europe - **Location**: ExCeL London, UK - **Dates**: 02.17 - 02.19, 2027 (3 Days) - **Description**: Europe's flagship technical AI conference returns to ExCeL London for three days, February 17-19, 2027 — bringing the world's leading AI labs, platforms, and tools to Europe's builders. - **Website**: [ai.engineer/europe](ai.engineer/europe) [Black and white photo of a grand building in London] [Black and white photo of Big Ben and Westminster Bridge] [Black and white photo of the London Eye] ### 00:16:20–00:16:40 # Stay in the Loop Scan the QR code or view the schedule at [ai.engineer/schedule](ai.engineer/schedule) [QR code linking to ai.engineer/schedule] ### 00:16:40–00:17:00 ## SILVER SPONSORS - Anchor - apify - AXD CAPITAL - Chainguard - CLOUDFLARE - DE Shaw & Co - Expo - extend - FLOX - Gravitee - Judgment Labs - Kimchi - MINIMAX - NIMBLE - Pydantic - RELAI - SENTRY - specific ai - supabase - SurrealDB - tetrate - UNION View Full Schedule at [ai.engineer/nyc/2026/schedule](ai.engineer/nyc/2026/schedule) [Logos of the silver sponsors] ### 00:17:00–00:17:20 # AI Engineer NEW YORK [https://ai.engineer](https://ai.engineer) ### 00:17:20–00:17:40 # WE'LL BE RIGHT BACK ### AI Engineer NEW YORK View Full Schedule at [ai.engineer/nyc/2026/schedule](ai.engineer/nyc/2026/schedule) [Illustration of the New York City skyline at night] ### 00:17:40–00:18:00 # AI Engineer NEW YORK ## WE'LL BE RIGHT BACK View Full Schedule at [ai.engineer/nyc/2026/schedule](ai.engineer/nyc/2026/schedule) [Illustration of the New York City skyline at night]