# AI and the Future of Fundamental Investing — Jack Woodruff, Vinoo Ganesh — session 2026-10-13T18:55:00.000Z → 2026-10-13T19:13:00.000Z _20 transcript lines · 53 slides · source: full recording_ ## Transcript **[00:00:00 · 0]** Through your inbox or messing with Jira in conference. And so when you have unlimited token, you are free to do that. You no longer have to save your tokens for the end of the quarter or, the most difficult task. So where do we stand? Well, open source is now a priority, a revolute. **[00:00:19 · 0]** The economics are favorable. The billing is flat. We don't have any bill shock any of the month. And the frontier is the exception. So obviously, we keep using code, exclude code. **[00:00:30 · 0]** The latest model is fantastic. Jimmy and I are going very well, very good as well. But we keep them as exception. So how do we get there? Some metrics that matter when you're self hosting model. **[00:00:47 · 0]** So to serve 2,000 engineers, we roughly need 56 b 200, and that's to host GLM 5.3. GLM is a big model, right, parameters. What you want to aim for is sub three second, attempt to first token. This is empirical evidence. As soon as you exceed five seconds, my slacking box starts to blow up, and so we know that's the limit. **[00:01:14 · 0]** In terms of throughput, it doesn't really matter as much, right? So we serve, in terms of interactivity, 200 tokens per second. People are fine. They don't need to go higher. The way people work has changed. **[00:01:25 · 0]** We no longer look at the stream of thinking tokens. People run tasks in parallel and check-in when the task is done. But if you don't have high throughputs, and you're an engineer at Revolut, you just connect after 5PM when a one is on GPU, and you can get 800 TPS. And then you know, it's a dream. In terms of cost, in terms of impact, self hosting GLM, or blended cost of token is 30¢ per million. **[00:01:50 · 0]** And so our invoice is one tenth of what we were spending on Entropic and OpenAI. So there is a lot of tuning that goes into running open source models. If you just get GPUs and upload weights on a GPU and you start doing some inference, you're going to get terrible performance. That's because your workload is going to be very dependent on the parameters that you set. So if you look all the way to the left of that chart, the blue line is going to be demand. **[00:02:26 · 0]** So that's request per second. That's going to be what's going to be the bottleneck for your latency for your GPUs. And the red line is what we're optimizing for. That's the time to first token. So when we started back in April, you could see records per second around three, which is high but not so high, and time to first token over ten seconds, which is very slow. **[00:02:50 · 0]** Part of the data that you see is the same number of GPU. We did an increase GPU count. We still have 56. And by the end, you see in August, requests per second were above over seven. And time to first token were around 1.6. **[00:03:04 · 0]** So we have had 88% improvement in performance for 76% increased demand. So the learning from this is there is a lot of parameters that you can improve on, and you really need to raise the scaling if you are looking to self host a model. So let's talk parameters. There's no secret recipe, but the things you should consider are the following. First one, long context is a scam. **[00:03:33 · 0]** You really need to be mindful that even though you can serve 1,000,000 tokens, two fifty six, two fifty six is really what you need. Contamination, if you're running black wall generation GPU, you don't want to run anything higher than FP four. And the FP4 is best. Roughly, going from FP4 to FP8 is a 40% saving in GPU footprint for zero loss in precision. Other parameters to consider tensor priotism. **[00:04:02 · 0]** Even though GLM fits on TP four, you don't have enough kv cache. And so maybe a TP eight works better, depending on how large your workload is. And then on the more zoterek, check your cache hit. Coding task needs to be above 90% cache hit. And you can look into optimize spec decoding, sequential pre fill, PDD sag, all of these. **[00:04:29 · 0]** Cool. So now let's talk about production setup. GMM obviously is a text only model, which is a limitation for some people. The trade off, I would say, we would make every time is get the smarter model and figure out how to do image recognition another way. You can easily have a local router, open Codexes one, where you would route any image recognition to Gemini or Luna and keep JN for the main task. **[00:04:57 · 0]** If you really need some vision model, JLAM five gs Flash works really well. We have our UI designer making great use of it. Turnaround for UI product has gone a really, really long way. And then the last one is going to be about the harness. So all of these endpoints are going to be OpenAI compatible API with response API. **[00:05:25 · 0]** And so you can use them inside open code, code codex. Codex app as well works very well. And so that's what we recommend. And then last message, this one is for the CIO in the room. There's been a lot of stigma in using Chinese open source model. **[00:05:50 · 0]** So the thing to consider is you're not using the inference dedicated endpoint that goes to China. You're just getting the weights and putting them on a US based service. So for data sharing side, this is totally fine. Revolut, in particular, because we're a EU regulated bank, we fall under the most strict regime in terms of compliance. And so what we find that works is all of these coding tasks are not considered autonomous agents. **[00:06:22 · 0]** They're always with a human in the loop. And so how do you give reassurance that you have a human in the loop? You start by creating a sandbox that guarantees that. And the sandbox means now if you've used code, it has all these features of approved for me. You just have to set the rules for the operation that you don't want to allow. **[00:06:42 · 0]** So you can allow grep but not Gradle, for instance. And that's going to give confidence to your regulator that it's not an autonomous agent that needs different supervision. The sandbox also protects you from rogue use. You may have bad actor inside your company, and you want to shield for that. You'd be surprised what you find in a large company. **[00:07:05 · 0]** And then the last one is prompt injection not so much a thing now given how smart models are, but always a thing. In terms terms of data privacy, we spoke about it. So as long as the data sits in one of the countries that are in the data equivalency framework, so US, EU, UK, Israel, you are fine. You can get your GPUs there. All of your inference provider or GPU provider are going to have zero data retention by default. **[00:07:34 · 0]** And then the last thing you need is a very strong open telemetry so that if you ever have an incident, you need to be able to go deep and look at the logs. It also gives you some analytics. All right. Thanks, everyone. That was all for my talk. ## Slides ### 00:00:00–00:00:20 # AI adoption & impact ## Where we started: - High skepticism of Frontier models ROI - Curious about open source ## Where we landed: - Opensource as priority: - Predictable invoice (no bill shock) - 60% lead time improvement ## Frontier as exception - Some tasks gpt/claude do best Shipping like a startup, regulated like a bank David Abitbol / Ops & Infrastructure Leader **Revolut** [Line graph titled "Lead time engineering task completion" showing two lines representing SMA3 Lead Time (Days) and SMA3 Coding Time (Hours) from January '25 to July '26, with annotations for Gemini web chatbot, Claude/Codex, and Open source LLM self hosted.] ### 00:00:20–00:00:40 # AI adoption & impact ### Where we started: - High skepticism of Frontier models ROI - Curious about open source ### Where we landed: - Opensource as priority: - Predictable invoice (no bill shock) - 60% lead time improvement - Frontier as exception - Some tasks gpt/claude do best ## Shipping like a startup, regulated like a bank David Abitbol / Ops & Infrastructure Leader Revolut [Line graph titled "Lead time engineering task completion" showing two lines (Lead Time in Days and Coding Time in Hours) decreasing over time from January 2025 to July 2026, with annotations for Gemini web chatbot, Claude/Codex, and Open source LLM.] ### 00:00:40–00:01:00 # AI adoption & impact ### Where we started: - High skepticism of Frontier models ROI - Curious about open source ### Where we landed: - Opensource as priority: - Predictable invoice (no bill shock) - 60% lead time improvement - Frontier as exception - Some tasks gpt/claude do best ## Shipping like a startup, regulated like a bank David Abitbol / Ops & Infrastructure Leader Revolut [Line graph titled "Lead time engineering task completion" showing two lines (Lead Time in Days and Coding Time in Hours) over time, with vertical markers for events like Gemini web chatbot, Claude/Codex, and Open source GLM] ### 00:01:00–00:01:20 # Metrics that matter for happy engineers - **2,000**: Engineers served with GLM 5.3 on 56 x B200s with Together; token guilt free, - **<3s**: TTFT; after 5s Slack messages begin to increase - **>200**: TPS target - **$0.30**: Per M token in/out (cache is free) 1/10th of the cost ### Shipping like a startup, regulated like a bank David Abitbol / Ops & Infrastructure Leader **Revolut** ### 00:01:20–00:01:40 # Metrics that matter for happy engineers - **2,000**: Engineers served with GLM 5.3 on 56 x B200s with Together; token guilt free, - **<3s**: TTFT; after 5s Slack messages begin to increase - **>200**: TPS target - **$0.30**: Per M token in/out (cache is free) 1/10th of the cost ## Shipping like a startup, regulated like a bank David Abitbol / Ops & Infrastructure Leader Revolut ### 00:01:40–00:02:00 # Metrics that matter for happy engineers - **2,000**: Engineers served with GLM 5.3 on 56 x B200s with Together; token guilt free, - **<3s**: TTFT; after 5s Slack messages begin to increase - **>200**: TPS target - **$0.30**: Per M token in/out (cache is free) 1/10th of the cost ## Shipping like a startup, regulated like a bank David Abitbol / Ops & Infrastructure Leader **Revolut** [Logo of three interconnected circles in the top right corner] ### 00:02:00–00:02:20 # Resist scaling, tune first 80% of the gain came from adjusting hyperparameters, not more GPUs - **76%** Higher RPS - Migration to stable, same 56x B200 GPUs throughout - **88%** Lower TTFT - From 14.26s mid-migration to 1.68s stable Also tuned: - Long context (256k-1M): costs KV cache, little gain - FP4 quantization - Tensor parallelism - Cache hits - Sticky sessions - New: prefill/decode (PD) disaggregation - On Together AI's Dedicated Model Inference ## Shipping like a startup, regulated like a bank David Abitbol / Ops & Infrastructure Leader Revolut [Line graph titled "TTFT versus load over time" showing TTFT (seconds) and Requests per second over a period from late June to late August, with annotations for different migration phases and stable setup.] ### 00:02:20–00:02:40 # Resist scaling, tune first 80% of the gain came from adjusting hyperparameters, not more GPUs - **76%** Higher RPS - Migration to stable, same 56x B200 GPUs throughout - **88%** Lower TTFT - From 14.26s mid-migration to 1.68s stable ### TTFT versus load over time Lower demand did not protect TTFT during the migration. The final setup handled higher demand with much lower TTFT. Also tuned: - Long context (256k-1M): costs KV cache, little gain - FP4 quantization - Tensor parallelism - Cache hits - Sticky sessions - New: prefill/decode (PD) disaggregation - On Together AI's Dedicated Model Inference [Line graph showing TTFT (seconds) and Requests per second over time, with phases labeled as TPB Baseline, TPB Migration, and TPB Stable] ### 00:02:40–00:03:00 # Resist scaling, tune first ## 80% of the gain came from adjusting hyperparameters, not more GPUs ### 76% Higher RPS - Migration to stable, same ### 00:03:00–00:03:20 # Resist scaling, tune first 80% of the gain came from adjusting hyperparameters, not more GPUs - **76%** Higher RPS - Migration to stable, same 5 ### 00:03:20–00:03:40 # Resist scaling, tune first 80% of the gain came from adjusting hyperparameters, not more GPUs - **76%** Higher RPS Migration to stable, same ### 00:03:40–00:04:00 # Resist scaling, tune first 80% of the gain came from adjusting hyperparameters, not more GPUs - 76% Higher RPS - Migration to stable, same 56x B200 GPUs throughout - 88% Lower TTFT - From 14.26s mid-migration to 1.68s stable ### TTFT versus load over time Lower demand did not protect TTFT during the migration. The final setup handled higher demand with much lower TTFT. Also tuned: - Long context (256k-1M): costs KV cache, little gain - FP4 quantization - Tensor parallelism - Cache hits - Sticky sessions - New: prefill/decode (PD) disaggregation - On Together AI's Dedicated Model Inference [Line graph showing TTFT (Time To First Token) and RPS (Requests Per Second) over time, with annotations for different migration phases and their impact on performance metrics.] ### 00:04:00–00:04:20 # Resist scaling, tune first 80% of the gain came from adjusting hyperparameters, not more GPUs - **76%** Higher RPS - Migration to stable, same 56x B200 GPUs throughout - **88%** Lower TTFT - From 14.26s mid-migration to 1.68s stable ### Also tuned: - Long context (256k-1M): costs KV cache, little gain - FP4 quantization - Tensor parallelism - Cache hits - Sticky sessions - New: prefill/decode (PD) disaggregation - On Together AI's Dedicated Model Inference --- Shipping like a startup, regulated like a bank David Abitbol / Ops & Infrastructure Leader Revolut [Line graph showing TTFT (Time To First Token) and RPS (Requests Per Second) over time, with annotations for migration phases and baseline/stable states] ### 00:04:20–00:04:40 # Resist scaling, tune first 80% of the gain came from adjusting hyperparameters, not more GPUs - **76% Higher RPS** Migration to stable, same 5 ### 00:04:40–00:05:00 # Production Setup ## Shipping like a startup, regulated like a bank ### Text models Written description in working code out ### Vision Models Mockup in working interface out. ### Harness / App CLI vs App ### 00:05:00–00:05:20 # Production Setup ### Text models - Written description in working code out ### Vision Models - Mockup in working interface out. ### Harness / App - CLI vs App ### 00:05:20–00:05:40 # Production Setup ### **Text models** - Written description in - working code out ### **Vision Models** - Mockup in - working interface out. ### **Harness / App** - CLI vs App ## Shipping like a startup, regulated like a bank David Abitbol / Ops & Infrastructure Leader Revolut ### 00:05:40–00:06:00 # For the CIO in the room When your codebase is subject to audit (and as a regulated bank) AI is scrutinized. Coding agents are considered "productivity use cases with humans in the loop," and don't need the same level of regulatory audit. For compliance, you need: ### A sandbox that: - Guarantees humans in the loop - Prevents rogue use - Protects from prompt injection ### Data privacy - Zero data retention - Data processing in U.S., EU, UK, ISL (data equivalency framework) ### Analytics & logs - For investigations & audits ## Shipping like a startup, regulated like a bank David Abitbol / Ops & Infrastructure Leader **Revolut ### 00:06:00–00:06:20 # For the CIO in the room When your codebase is subject to audit (and as a regulated bank) AI is scrutinized. Coding agents are considered "productivity use cases with humans in the loop," and don't need the same level of regulatory audit. For compliance, you need: - **A sandbox that:** - Guarantees humans in the loop - Prevents rogue use - Protects from prompt injection - **Data privacy** - Zero data retention - Data processing in U.S., EU, UK, ISL (data equivalency framework) - **Analytics & logs** - For investigations & audits ## Shipping like a startup, regulated like a bank David Abitbol / Ops & Infrastructure Leader **Revolut** [3D cube icon next to "A sandbox that:"] [Padlock icon next to "Data privacy"] [Two connected people icon next to "Analytics & logs"] ### 00:06:20–00:06:40 # For the CIO in the room When your codebase is subject to audit (and as a regulated bank) AI is scrutinized. Coding agents are considered "productivity use cases with humans in the loop," and don't need the same level of regulatory audit. For compliance, you need: **A sandbox that:** - Guarantees humans in the loop - Prevents rogue use - Protects from prompt injection **Data privacy** - Zero data retention - Data processing in U.S., EU, UK, ISL (data equivalency framework) **Analytics & logs** - For investigations & audits --- ### Shipping like a startup, regulated like a bank David Abitbol / Ops & Infrastructure Leader Revolut [Icon of a 3D cube] [Icon of a padlock] [Icon of connected people] [Small logo of three overlapping colored circles in the top right corner] ### 00:06:40–00:07:00 # For the CIO in the room When your codebase is subject to audit (and as a regulated bank) AI is scrutinized. Coding agents are considered "productivity use cases with humans in the loop," and ### 00:07:00–00:07:20 # For the CIO in the room When your codebase is subject to audit (and as a regulated bank) AI is scrutinized. Coding agents are considered "productivity use cases with humans in the loop," and don't need the same level of regulatory audit. For compliance, you need: ### A sandbox that: - Guarantees humans in the loop - Prevents rogue use - Protects from prompt injection ### Data privacy - Zero data retention - Data processing in U.S., EU, UK, ISL (data equivalency framework) ### Analytics & logs - For investigations & audits ## Shipping like a startup, regulated like a bank David Abitbol / Ops & Infrastructure Leader **Revolut** [Icon of a cube] [Icon of a padlock] [Icon of connected people] [Small logo of three overlapping colored circles] ### 00:07:20–00:07:40 # For the CIO in the room When your codebase is subject to audit (and as a regulated bank) AI is scrutinized. Coding agents are considered "productivity use cases with humans in the loop," and don't need the same level of regulatory audit. For compliance, you need: ### A sandbox that: - Guarantees humans in the loop - Prevents rogue use - Protects from prompt injection ### Data privacy - Zero data retention - Data processing in U.S., EU, UK, ISL (data equivalency framework) ### Analytics & logs - For investigations & audits ### Shipping like a startup, regulated like a bank David Abitbol / Ops & Infrastructure Leader **Revolut** [Icon of a cube above "A sandbox that:"] [Icon of a padlock above "Data privacy"] [Icon of connected people above "Analytics & logs"] [Small logo of three overlapping circles in the top right corner] ### 00:07:40–00:08:00 # For the CIO in the room When your codebase is subject to audit (and as a regulated bank) AI is scrutinized. Coding agents are considered "productivity use cases with humans in the loop," and don't need the same level of regulatory audit. For compliance, you need: **A sandbox that:** - Guarantees humans in the loop - Prevents rogue use - Protects from prompt injection **Data privacy** - Zero data retention - Data processing in U.S., EU, UK, ISL (data equivalency framework) **Analytics & logs** - For investigations & audits ## Shipping like a startup, regulated like a bank David Abitbol / Ops & Infrastructure Leader Revolut [Icon of a cube] [Icon of a padlock] [Icon of connected people] [Small logo of three overlapping colored circles] ### 00:08:20–00:08:40 # 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 illuminated buildings and a bridge in the distance] ### 00:08:40–00:09:00 # AI Engineer Europe 2027 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. **Location:** ExCeL London, UK **Dates:** 02.17 - 02.19, 2027 (3 Days) **Website:** [ai.engineer/europe](ai.engineer/europe) [Black and white photo of the London Eye] [Black and white photo of Big Ben and Westminster Bridge] [Black and white photo of a grand building in London] ### 00:09:00–00:09:20 # 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] ### 00:09:20–00:09:40 # AI Engineer ## NEW YORK https://ai.engineer ### 00:09:40–00:10:00 # Stay in the Loop Scan the QR code or view the schedule at [ai.engineer/schedule](https://ai.engineer/schedule) [QR code linking to https://ai.engineer/schedule] ### 00:10:00–00:10:20 View Full Schedule at [ai.engineer/nyc/2026/schedule](ai.engineer/nyc/2026/schedule) ### 00:10:20–00:10:40 # SUPPORTING PARTNERS - ANTHROPIC - McKinsey & Company - Parasail - Cerebras - CTC - LaunchDarkly - parallel - Primitive - APOLLO - ARCA - axim - Basis - Bloomberg - BRIDGEWATER - CCAVENTUS - COATUE - coinbase - imec - Compound - COVAL - epsilab - featherless.ai - Fieldguide - general intuition - JELLYFISH - kaggle - Linqalpha - omni - OpenProse - ramp - Revolut - rogo - Rowspace - runway - şamaya - sfcompute - together.ai - TWO SIGMA View Full Schedule at [ai.engineer/nyc/2026/schedule](ai.engineer/nyc/2026/schedule) ### 00:10:40–00:11:00 # The Highest-Signal Technical AI Conferences AI Engineer hosts AI developer conferences around the world for engineers, breakout startup founders, and AI architects advancing AI engineering at the world's leading technology companies. ## Find Out What's Next Scan the QR code or visit https://ai.engineer to subscribe to our newsletter and stay up to date with the latest news and events. [QR code linking to https://ai.engineer] ### 00:11:00–00:11:20 # 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 of San Francisco with the Bay Bridge in the background] ### 00:11:20–00:11:40 # AI Engineer Europe 2027 ## ExCeL London ### 02.17 - 02.19 2027 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. [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 and River Thames] ### 00:11:40–00:12:00 # AI Engineer NEW YORK ## WE'LL BE RIGHT BACK View Full Schedule at [ai.engineer/nyc/2026/schedule](ai.engineer/nyc/2026/schedule) [Silhouette of the New York City skyline at night, featuring landmarks like One World Trade Center, the Empire State Building, and the Statue of Liberty] ### 00:12:00–00:12:20 # 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] ### 00:12:20–00:12:40 # WE'LL BE RIGHT BACK ## AI Engineer NEW YORK View Full Schedule at [ai.engineer/nyc/2026/schedule](ai.engineer/nyc/2026/schedule) [Nighttime skyline of New York City with prominent skyscrapers and the Statue of Liberty] ### 00:12:40–00:13:00 # 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 of San Francisco with a bridge in the background] [G2i logo] ### 00:13:00–00:13:20 # AI Engineer Europe 2027 ## ExCeL London ### 02.17 - 02.19 2027 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. [ai.engineer/europe](ai.engineer/europe) [Black and white photos of London landmarks including the London Eye, Big Ben, and a historic building] ### 00:13:20–00:13:40 # PRESENTED BY ## Arize View Full Schedule at [ai.engineer/nyc/2026/schedule](ai.engineer/nyc/2026/schedule) [Arize logo, a pink upward-pointing caret shape] ### 00:13:40–00:14:00 # AI Engineer ## NEW YORK [https://ai.engineer](https://ai.engineer) ### 00:14:00–00:14:20 ## AI Engineer NEW YORK # Stay in the Loop Scan the QR code or view the schedule at [ai.engineer/schedule](https://ai.engineer/schedule) [QR code linking to https://ai.engineer/schedule] ### 00:14:20–00:14:40 ## 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] ### 00:14:40–00:15:00 # SUPPORTING PARTNERS - ANTHROPIC - McKinsey & Company - Parasail - Cerebras - CTC - LaunchDarkly - parallel - Primitive - APOLLO - ARCA - axim - Basis - Bloomberg - BRIDGEWATER - CCAVENTUS - COATUE - coinbase - imec - Compound - COVAL - epsilab - featherless.ai - Fieldguide - general intuition - JELLYFISH - kaggle - Linqalpha - nous - omni - OpenProse - ramp - Revolut - rogo - Rowspace - runway - şamaya - sfcompute - togetherai - Towards AI - TWO SIGMA View Full Schedule at [ai.engineer/nyc/2026/schedule](ai.engineer/nyc/2026/schedule) ### 00:15:00–00:15:20 # WE'LL BE RIGHT BACK ## AI Engineer NEW YORK View Full Schedule at [ai.engineer/nyc/2026/schedule](ai.engineer/nyc/2026/schedule) [New York City skyline at night] ### 00:15:20–00:15:40 # AI Engineer Code Summit PRESENTED BY **Google DeepMind** NOVEMBER 10-12, 2026 SF // CA AIE CODE returns for its second ### 00:15:40–00:16:00 # AI Engineer Europe 2027 ## AI Engineer Europe - **Location**: ExCeL London, London, UK - **Dates**: February 17-19, 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 the London Eye] [Black and white photo of Big Ben and Westminster Bridge] [Black and white photo of a classical building on a London street] ### 00:16:00–00:16:20 # AI Engineer NEW YORK ## WE'LL BE RIGHT BACK View Full Schedule at [ai.engineer/nyc/2026/schedule](ai.engineer/nyc/2026/schedule) [Silhouette of the New York City skyline at night, including One World Trade Center, the Empire State Building, and the Statue of Liberty] ### 00:16:20–00:16:40 # AI Engineer ## NEW YORK # WE'LL BE RIGHT BACK View Full Schedule at [ai.engineer/nyc/2026/schedule](ai.engineer/nyc/2026/schedule) [Silhouette of the New York City skyline at night, including the Statue of Liberty, One World Trade Center, and the Empire State Building] ### 00:16:40–00:17:00 # AI Engineer ## NEW YORK [https://ai.engineer](https://ai.engineer) ### 00:17:00–00:17:20 # AI Engineer Code Summit **NOVEMBER 10-12, 2026** **SF // CA** PRESENTED BY **Google DeepMind** 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 illuminated buildings] ### 00:17:20–00:17:40 # AI Engineer Europe 2027 ## ExCeL London 02.17 - 02.19 2027 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. [ai.engineer/europe](ai.engineer/europe) [Black and white photo of a classical building in London] [Black and white photo of Big Ben and the Houses of Parliament over a bridge] [Black and white photo of the London Eye] ### 00:17:40–00:18:00 ### PRESENTED BY arize View Full Schedule at ai.engineer/nyc/2026/schedule [Arize logo, a stylized pink upward-pointing arrow or house shape]