NDC London 2026
2026-01-26..2026-01-30
-- online --
[NDC London 2017, NDC London 2018, NDC London 2019, NDC London 2020, NDC London 2021, NDC London 2023, NDC London 2024, NDC London 2025, NDC London 2026]
who it's for
software developers, ai engineers, platform engineers, backend developers, frontend developers, site reliability engineers, engineering managers
topics
genetic algorithms for self-improving systems; prompt and context engineering; vector search with OpenSearch; embeddings in PostgreSQL with pgvector; disaggregated LLM inference architecture; observability for LLM latency and failures; passkeys and WebAuthn sign-in flows; .NET game emulation and MonoGame; WebRTC voice UIs with Twilio and Deepgram; accessibility and WCAG remediation; OpenTelemetry and OpAMP at scale
| title | author | length |
|---|---|---|
| Solutions That Evolve: Building Self-Improving Systems with Genetic Algorithms | Barry Stahl | 57m |
Genetic algorithm encodes a Shoots-and-Ladders strategy as a 299-gene chromosome (679 total alleles) producing about 10^103 possible strategies, runs six… | ||
| Flying a drone with twitter, bananas & Web APIs | Nhlanhla Lucky Nkosi | 38m |
Five concrete hacks to control a toy drone: network sniffing of a cheap SG900X fails so a Parrot Mambo Mini is bought instead; BLE and Web Bluetooth are used… | ||
| Trapped in the AI Error Loop: Why Smart Systems Keep Making Dumb Mistakes | Omphile Matheolane | 40m |
Explains why accepting a Copilot or ChatGPT suggestion can seed future failures as accepted fixes become the model’s context, producing repeating bugs, brittle… | ||
| The New Frontend Toolkit: Popover API, Dialog, and Next-Gen CSS | Eric Veliyulin | 48m |
Live-coding shows native dialog element and Popover API used with plain HTML/CSS/JS: calls to dialog.showModal() and dialog.close(), popover with… | ||
| Coding 4 Fun: 8-bit game emulation in .NET | Alex Thissen | 01h 04m |
A working Atari Lynx emulator written in C# on .NET and using MonoGame runs cross‑platform and models the 6502/65CO2 CPU plus Suzie and Mikey custom chips… | ||
| Hello JARVIS - Building Voice Interfaces for Your LLMS | Nathaniel Okenwa | 49m |
Live demo builds a Jarvis‑like voice UI that streams WebRTC audio via Twilio + Deepgram + OpenAI + 11Labs, pauses playback on new utterances, sends transcripts… | ||
| How to do half-assed UX | Ash Banaszek | 39m |
A Fortune 150 rollout of a third-party time system goes live to 32,000 employees with only 10‑minute training videos, producing 30+ minute hold times, disabled… | ||
| Going Passwordless - A Practical Guide to Passkeys in ASP.NET Core | Maarten Balliauw | 56m |
Passkeys replace passwords using the WebAuthn protocol and demonstrations use WebAuthn with navigator.credentials flows, a Blazor app on .NET 10 and ASP.NET… | ||
| Stop Treating LLMs Like REST APIs | Jeff Fran, Jack Pearce | 42m |
Claims that treating LLM inference like REST causes launch-day failure, and shows a concrete fix: deploy LLMD on Kubernetes with disaggregated prefill vs… | ||
| Postgres and OpenAI - a match made in Azure | Giorgi Dalakishvili | 44m |
Shows PostgreSQL on Azure calling Azure OpenAI via the azure_ai extension to generate embeddings, then stores them with pgvector and a Microsoft research kN… | ||
| Accessibility by Everyone (and for Everyone) | Amy Kapernick | 50m |
Audit data shows the top 1,000,000 homepages have 50,960,288 accessibility issues (≈51 per site) and 94.8% fail basic standards; 96% of those issues are easy… | ||
| RolePl-AI | Jo Franchetti | 51m |
Prototype blends rulebound procgen with local LLMs to produce coherent RPG universes: a three‑pass pipeline (outline, detail, polish) emits strict JSON schemas… | ||
| Optimizing your HttpClient usage | Nico Vermeir | 01h 00m |
Shows that creating 20,000 new HttpClient instances exhausts sockets after a 2s API delay and explains why reusing connections matters: DNS, TCP handshake and… | ||
| Application Performance Optimisation in Practice | Steve Gordon | 01h 02m |
Parses and sanitises SQL statements for OpenTelemetry, then iteratively targets hotspots using dotTrace, dotMemory and BenchmarkDotNet. Replaces expanding… | ||
| Blazor Authentication? Hold My Beer! | Roland Guijt | 01h 00m |
Explains concrete recipes for cookie + OpenID Connect in Blazor Server, WebAssembly and Auto (hybrid): use authorization code flow with a server-side token… | ||
| The 1 Trillion Pixel Maze: My Decade-Long Obsession with C# Optimization | Davy Davidse | 49m |
C# implementation rethinks maze generation end to end: backtrack algorithm with bit arrays, disk‑tiled caching and chunked enumerable processing yields huge… | ||
| Getting maximum mileage from a minimal number of tests | Andrew Poole | 54m |
Shows how one automated end-to-end component test runs three executables in-memory so a single test covers the full happy path across API, event listener and… | ||
| Code that writes code - .NET Source generators | Glenn F. Henriksen | 55m |
Source generators hook into the Roslyn pipeline (must target .NET Standard 2.0) to produce ephemeral .g.cs files at compile time, removing reflection and… | ||
| OpenTelemetry At Scale 101: Intro to OpAMP | Aakansha Priya, Adriana Villela | 52m |
OpAMP is a protocol to centrally manage fleets of OpenTelemetry Collectors across clouds and edges, solving config drift and risky rollouts for hundreds to… | ||
| HTMX & ASP.NET Razor Pages – Web development without the JavaScript pains | Kevin DeRudder | 58m |
Builds a working playlist/memories app using HTMX with ASP.NET Razor Pages, Bulma for styling and optional Hyperscript and SweetAlert for UX; server renders… | ||
| A Defence of Technical Excellence | Chris Simon | 51m |
Argues that internal quality drives a reinforcing loop of confusion, bugs and slowed delivery, and models that loop with causal loop diagrams using the browser… | ||
| How to Git Away with Murder | Sergès Goma | 44m |
Claim: Git stores immutable snapshots in a content-addressed object DB and branches are just pointers to commit hashes. Practical tools demonstrated: feature… | ||
| Meditations on Code As Art | David Whitney | 50m |
Shows Tracy Emmen’s My Bed (sold for £2.54M) as proof that messy, autobiographical objects can be art and compares that to software artifacts like the Apollo… | ||
| If only I had known ... | Hannes Lowette | 48m |
Practical, actionable career habits: adopt a three-to-five year plan and break it into tiny, repeatable steps using the visualize 0-to-10 progress exercise… | ||
| Digging into the Matrix: Practicing Code Archaeology | Arthur Doler | 01h 02m |
A practical five-step method—mindset, surveying, excavation, analysis, progression—shows how to understand and change code you didn’t write. It uses concrete… | ||
| Building AI That Learns and Adapts: A Case Study in MRI Diagnostics | Agata Chudzińska | 43m |
201 MRI MRI series (plus 100 normals and 50 other pathologies) train a U-Net style segmentation pipeline; initial model detects about 70% of FCDs but produces… | ||
| Crafting Intelligent Agents with Context Engineering | Carly Richmond | 57m |
Defines context engineering as assembling the exact data you pass into an LLM and shows a working travel agent called Sorly travel bot built with TypeScript AI… | ||
| Building Secure Infrastructure for Productive AI Agents | Eric Paulsen, Jiachen Jiang | 01h 03m |
Claims that agentic AI can wreck production are backed by a concrete Replit incident where a user loses a production codebase, and by experiments with… | ||
| Between the Layers– Interpreting Large Language Models | Michelle Frost | 01h 01m |
Survey of recent interpretability work argues that mechanistic features replace single-neuron stories: Anthropic’s 2023–2024 papers use sparse autoencoders to… | ||
| From Model Integration to Agents: Building Intelligent .NET Applications | Gill Cleeren | 01h 08m |
Walks through concrete .NET integrations: uses the OpenAI.NET SDK for chat, streaming and image bytes (images take ~30–40s), Azure AI Services for language… | ||
| How I used AI voice to help me play Magic: The Gathering | Michelle "MishManners" Duke | 01h 05m |
Built an AI voice assistant that answers Magic: The Gathering rules questions by piping Twilio voice through a websocket to a Python app, calling Twilio… | ||
| AI Agents Need Permission Slips | Heather Downing | 52m |
A concrete blueprint for safe agents: an agent at Replet in July 2025 executes a drop database and wipes ~1,190 customer accounts, prompting mandates for… | ||
| Vector Search Made Simple: Getting Started with OpenSearch for AI Applications | Dotan Horovits | 55m |
OpenSearch claims end-to-end vector search at production scale: 140 GitHub repos, 1.6 billion downloads, and 3,000 contributors, and it runs billions to… | ||
| Modular Code with Examples in C# | Ian Cooper | 01h 07m |
Argues that in C# the useful module boundary is the assembly not namespaces or folders, demonstrates a small Quest Worlds contest resolver implemented as… | ||
| "Run Query Run" - A Fresh Look at SQL Wait Stats | Pinal Dave | 58m |
Live demos extract wait stats from SQL Server, show a parallel query eating CPU but producing only 75s of useful work while CXPACKET waste = 389s wasted, and… | ||
| Advanced Pattern Matching in C# | Oliver Sturm | 01h 02m |
Walks through concrete C# pattern features added in C# 7–11 pattern features (type, property, positional/tuple, relational, list and spread patterns), shows… | ||
| Notebooks & .NET: A Marriage Made In Jupyter | Alexander Hall | 58m |
Explains how to install the global tool dotnet-interactive and use the .NET kernel inside polyglot notebooks so cells can run C#, F#, PowerShell and share… | ||
| Modern .NET Configuration Practices | Chris Ayers | 01h 01m |
Covers concrete practices: use JSON, environment variables, command-line, Azure App Config and Key Vault as configuration providers; bind hierarchical settings… | ||
| 10 Tips To Level Up Your AI-Assisted Coding | Aleksander Stensby | 01h 02m |
Claim: recent models cross a usability threshold — Opus 4.5 marks a before/after where agents go from useless to writing production features, with anecdotes of… | ||
| Reliable Agentic Systems need Durable Execution 🤖 | Marc Duiker | 01h 00m |
Claim: durable execution via Dapr Workflow lets long‑running agentic pipelines resume from checkpoints and avoid re‑calling LLMs. Live demos run locally with a… | ||
| Keynote: AI-Powered App Development | Steve Sanderson | 54m |
Demo shows Copilot CLI adding a new flag and implementation (the --show-models in ~1 minute example) and running parallel reviews across models… | ||
| Think like a user: practical UX design tips for developers | Lex Lofthouse | 57m |
Practical rules: assume users scan pages in under 3 seconds following an F-shaped reading pattern, so give a clear starting point, tight visual hierarchy… | ||
| Let's break some WCAG rules | Elise Kristiansen | 53m |
15–25% of people need accommodations and 94.8% of websites fail at least one WCAG AA rule: 79.8% bad contrast, 55% missing alt text, 48% missing labels, 45%… | ||
| Usability Testing is easier than you think | Jo Minney | 46m |
Argues that small qualitative usability tests beat big quantitative ones for fast fixes: run 5 users with simple tasks, record sessions, coach think‑aloud… | ||
| Prompt Failures and Latency Spikes: Observability for AI | Prerit Munjal | 01h 00m |
Demonstrates deploying OpenLit with Langfuse and a proxyed light LLM to capture trace-based prompt chains, surface per-call token counts and cost, and enforce… | ||
| Social Engineering: Hacking Humans | Pawel Sucholbiak | 58m |
Lists breaches (Facebook $99M, Ubiquiti $39.1M, Toyota $37M) and a supply‑chain incident that compromises 18 npm packages, two billion downloads via a backdoor… | ||
| Correcting Common Async/Await Mistakes in .NET 10 | Brandon Minnick | 01h 05m |
Shows that the C# compiler generates an IAsyncStateMachine (usually a struct for release builds), wraps moveNext in a giant try/catch so un-awaited tasks hide… | ||
| Quantum Physics for Software Developers | Filip W | 01h 00m |
Code reproduces beam-splitter and Mach–Zehnder results in Q#: a single qubit with an H gate and measurement yields 1,000 runs ≈ 50/50, two H gates yield 1,000… | ||
| Language Games | Eli Holderness | 47m |
Argues that communication is a sequence of four stages from idea to perception and that clarity comes from matching signifiers to concepts using semiotics… | ||
| Building a Carbon Impact Agent: Boosting ROI with Sustainable AI in... | Darshna Shah | 51m |
Shows a working prototype Carbon Impact Agent built as a Langraph agent that ingests per-request telemetry (tokens, duration, region, task) and uses a carbon… | ||
| Open Source is Broken, and it’s Our Fault | Lotte Pitcher | 49m |
Many critical products rest on unpaid maintainers, with high‑impact failures like Heartbleed ~$500M cost and the log4j weekend firefight showing systemic risk… | ||
| neo4j for the relational .NET developer | Chris Klug | 01h 03m |
Neo4j proves practical for .NET devs: local runs in 3 minutes to run locally via Docker or Neo4j Desktop, or use Aura DB free tier in the cloud; the movie… | ||
| Learning not leaning: How developers can Accelerate their Learning with AI | Emma Burstow | 56m |
Umbraco CMS team experiments over a year turn AI from a crutch into a learning scaffold: a mandatory Copilot-assisted sprint with 53 engineers (about one third… | ||
| Playing the long game | Sheena O'Connell | 46m |
Argues that LLM progress is messy not linear: scaling laws conflict, inference tricks matter, and DeepSeek R1 trains for about a tenth of GPT‑4 cost, which… | ||
| Resilient by Design | Chris Ayers | 58m |
Argues that resilience starts with concrete numbers and policies — SLIs/SLOs, error budgets, RPO/RTO — and operational practices like load testing, chaos game… | ||
| The Fun Side of Advanced TypeScript: An Interactive Coding Session | Dante De Ruwe | 54m |
Builds a working Wordle fully in TypeScript types (about 21 lines) and a type-level calculator that does increment, decrement, add, multiply, whole division… | ||
| Indiana Jones and the Temple of Legacy Code | William Brander | 51m |
Gives concrete tactics: establish reproducible builds (lock deps, build.sh/build.ps1, devcontainers), push that to CI, and add at least one test (Playwright if… | ||
| What in the Hunger Games is Happening with Recruitment? | Suzi Edwards Alexander | 50m |
A rapid history and prescriptive playbook: hiring evolves from Edison’s 146‑question gatekeeping with a 2% pass rate to LinkedIn in 2002 that productizes… | ||
| The Great Brain Robbery: Navigating the Dark Future of Online Manipulation | Jeff Watkins | 51m |
A 2024 deepfake video call causes a $25 million deepfake heist by impersonating an executive; attackers scale via generative models and dark‑web tools like… | ||
| Supercharged Testing: AI-Powered Workflows with Playwright + MCP | Debbie O'Brien | 48m |
Demonstrates generating end to end tests by wiring a local LLM to the Playwright MCP Test Server and running three specialized agents planner/generator/healer… | ||
| Anatomy of an Incident | Liam Westley | 48m |
A bad CrowdStrike update ships an old data file so a kernel‑mode C++ routine expects 21 vs 20 array items, triggers blue screens across Windows hosts for about… | ||
| Bringing stories to life with AI, data streaming and generative agents | Olena Kutsenko | 41m |
Claims that LLM-driven agents can form coherent societies with memory and reflection, demonstrated by a Stanford demo of ~30 characters where over 50% of… | ||
| Drop the Bass with Embedding and Vectors in Azure AI Search | Alan Smith | 01h 01m |
Live demos show Azure AI Search doing text, image, face and music vector search by indexing embeddings rather than keywords: uses Azure AI Search, ResNet-18 +… | ||
| .NET Meets MCP: Build Your Own AI-Powered Service with C# | Gerald Versluis | 01h 01m |
Explains how to run and build MCP servers with C# and .NET, showing live VS Code demos that register tools, expose prompts and resources, and run as stdio or… | ||
| Building Identity into LLM Workflows with Verifiable Credentials | Ben Dechrai | 01h 02m |
A security researcher social engineers Claude (June 2025) until the model drafts and wants to run a jailbreak implementation, then demonstrates an… | ||
| Supercharging Local Development with Aspire | Jimmy Bogard | 01h 01m |
Aspire proves to cut “time‑to‑login‑screen” by replacing ad hoc scripts and Docker Compose with a .NET 10 app host that orchestrates projects, Docker (OCI)… | ||
| Beyond the AI Hype: What's Real, What's Next | Richard Campbell | 52m |
Generative models actually scale into products: a token-based GPT lineage goes from GPT‑3 at 175 billion parameters to multi‑modal GPT‑4 and an optimized… | ||
| How we support multi-cloud apps at Stack Overflow | Chris O'Dell | 34m |
Stack Overflow runs Teams and Enterprise on Azure and the public site on GCP, so the platform team builds a layer that hides cloud differences using… | ||
| How I Tamed Claude | Emmz Rendle | 01h 06m |
A concrete workflow for getting production code from Claude Code: use a project claude.md plus system prompts for four roles—analyst, architect, developer… | ||
| How I used AI voice to help me play Magic: The Gathering | Michelle "MishManners" Duke | 01h 04m |
Built a voice assistant that answers Magic the Gathering rules and guides play by piping speech-to-text → LLM → text-to-speech, using Twilio Conversation Relay… | ||
| Immutable Patterns of System Design; From Monolith to Agentic AI | James Eastham | 58m |
Argues five concrete design patterns—evolvability, encapsulation, coupling, understanding and designing for failure—and shows an agent added to a pizza… | ||
| The Undersea Cable Network | Richard Campbell | 51m |
About 1,700 operational undersea cables carry most global internet traffic and modern fiber builds like the 6,600 km Dant cable hold ~250 Tb using 12 fibers… | ||
| Java Sucks (So C# Didn't Have To) | Adele Carpenter | 58m |
Java explains why generics use type erasure (JSR-14 chose backward compatibility over VM changes), so runtime type info is gone and libraries must pass type… | ||
| Modular Monoliths and Other Facepalms | Kevlin Henney | 01h 04m |
Argues that modular monolith is rebranding rather than a new style, tracing modularity back to the 1960s and showing the same principles reappear; documents… | ||
| Do It With Style: Rethinking CSS | Dylan Beattie | 57m |
Shows how to replace common JS widgets with real HTML and modern CSS: an details/summary accordion becomes mutually exclusive via the name attribute (landed… | ||
| Beyond Pub/Sub - Advanced Messaging Patterns | Poornima Nayar | 01h 01m |
Explains concrete fixes: Transactional Outbox + database-stored outgoing messages plus handler-side deduplication and idempotency yields exactly-once… | ||
| Keynote: The dangers of probably-working software | Damian Brady | 55m |
Huffman compression code copied from Wikipedia fails in production because padding zeros collide with prefix codes and TIFFs already-compressed images explode… | ||
| nothing matches | ||