Sergio Sisternes
Microsoft MVP Developer Technologies - Technology Solutions Director - Azure, AI and GitHub Europe at EMEA
London, United Kingdom
Actions
Sergio is a Microsoft MVP in Developer Technologies and Director at EPAM's Microsoft Business Group, where he is helping clients to create AI Native SDLC platforms and accelerate Azure Application Modernisations programmes with Microsoft and GitHub technologies.
With 16+ years in software engineering, cloud architecture, and IT consulting, Sergio is also a Microsoft and GitHub Certified Professional, active Open Source contributor (microsoft/apm) and frequent speaker in events such as: GitHub Universe (United States), FOSDEM (Belgium), Google Developer Day, Open Source in Finance and numerous enterprise events for clients across EMEA.
Area of Expertise
Topics
Scaling SWE Agents: Manage Agent Primitives Like NPM Packages with Agent Package Manager (APM)
Managing SWE agents at scale is painful: teams copy-paste markdown files (prompts, instructions, skills) across repos, leading to chaos, inconsistencies, and zero reusability.
Enter Agent Package Manager (APM) — the open-source tool from Microsoft that brings real package management to AI agent primitives.
Declare dependencies in a simple apm.yml file → apm install resolves direct + transitive deps → agents auto-configure for GitHub Copilot, Cursor, Claude, Codex, Gemini and more.
In this session, discover how to treat agent primitives (Instructions, Skills, Prompts, Agents, MCP Servers) as shareable, versioned packages — unlocking enterprise-grade collaboration, consistency, and an internal agent library that actually scales.
- Target audience: Engineering leads, AI platform teams, developers building with agentic tools (Copilot, Claude, Cursor etc.)
- Level: 200 (familiarity with AI coding agents helpful)
- Duration: 45–60 minutes (including Q&A) > 25-30 minutes version available
GitHub Enterprise AI-Native SDLC: Secure, Governed, and Powered by the Full GitHub Platform
In 2026, AI agents and multi-agent systems are automating 30–60% of the SDLC, delivering 10–35% velocity gains—but only if governed properly. Shadow agents, context pollution, IP leaks, and compliance gaps can derail adoption.
This hands-on talk shares practical enterprise patterns to capture AI's upside while enforcing security and governance:
- Agent Package Manager (APM): Microsoft's open-source dependency manager for primitives (instructions, skills, prompts, agents, MCP servers). We refactored a production mono-repo (118+ customizations across 15+ tech domains) using thin routing instructions + on-demand skills + generic agents → 63% less always-on context noise, 60.6% efficiency, zero duplication.
- Enterprise RAG Blueprint: Multimodal, permissioned knowledge grounding to eliminate hallucinations.
- Secure Hosted Execution: Private VPC/Zero Trust compute + networking for Copilot/agents.
Tied together by the full GitHub platform (Copilot Enterprise, Actions + Agentic Workflows preview, Advanced Security, Code Quality, MCP policies).
Audience will leave with three key takeaways: standardize primitives with APM, leverage GitHub's end-to-end stack, and pilot governance for a smooth AI-native dev experience. Interactive pulse questions included to engage engineers, managers, and directors alike.
Preferred duration: 60 minutes (20 min talk + 30 minutes demo + 10 min Q&A)
Target audience: Mixed – Engineers (technical depth on primitives/APM/RAG), Management/Directors (business impact, risk reduction, velocity gains)
Technical level: Intermediate to advanced (assumes familiarity with GitHub Copilot/Agents; no live demo required)
Technical requirements: Standard presentation setup. Internet access.
Open Source in Finance Forum London 2026 Sessionize Event Upcoming
Global Azure London 2026 Sessionize Event
DevFest @ Bletchley Park 2025 Sessionize Event
Global AI Bootcamp - London 2025 Sessionize Event
South Coast User Group User group Sessionize Event
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top