You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
> **Hackathon Award Category: Offline-Ready AI Award**
21
21
>
22
-
> As part of the Build-a-thon Hack!, we have a special award category that will recognize the best performing AI solution with standout offline capabilities (local inference).
22
+
> This quest mapped to our Offline-Ready AI Award category, which recognized the best performing AI solutions with standout offline capabilities (local inference).
23
23
>
24
-
> Consider building an app that:
24
+
> Winning solutions featured apps that:
25
25
>
26
-
> -Processes sensitive data entirely on-device.
27
-
> -Uses Foundry Local for reasoning and a cloud storage service for optional sync or analytics.
26
+
> -Processed sensitive data entirely on-device.
27
+
> -Used Foundry Local for reasoning and cloud storage services for optional sync or analytics.
28
28
>
29
-
> Highlight in your submission how you:
29
+
> Winners demonstrated:
30
30
>
31
-
> -Achieve **privacy** (no sensitive data leaving the device).
32
-
> -Optimize for **latency** using local inference.
31
+
> -**Privacy** (no sensitive data leaving the device).
32
+
> -**Latency** optimization using local inference.
33
33
>
34
34
> Checkout some Foundry Local + JavaScript projects: -
@@ -24,19 +24,19 @@ The **AI Toolkit** is a powerful VS Code extension that streamlines generative A
24
24
>
25
25
> **Hackathon Award Category: Agent Architecture Award**
26
26
>
27
-
> As part of the Build-a-thon Hack!, we have a special award category that will recognize the best AI agents with innovative and well-architected designs.
27
+
> This quest mapped to our special award category that recognized the best AI agents with innovative and well-architected designs.
28
28
>
29
-
> Consider building:
29
+
> Winning submissions featured:
30
30
>
31
-
> - Agents that solve complex problems with multiple steps of reasoning
32
-
> - Agents that integrate with custom MCP servers to access specialized tools or data
33
-
> - Agents with sophisticated evaluation techniques that push performance boundaries
31
+
> - Agents that solved complex problems with multiple steps of reasoning
32
+
> - Agents that integrated with custom MCP servers to access specialized tools or data
33
+
> - Agents with sophisticated evaluation techniques that pushed performance boundaries
34
34
>
35
-
> Highlight in your submission how you:
35
+
> Winners demonstrated:
36
36
>
37
-
> -Demonstrate innovative agent design patterns
38
-
> - Leverage MCP servers for specialized tool integration
39
-
> -Showcase evaluation methods that validate agent performance
37
+
> -Innovative agent design patterns
38
+
> - Leverage of MCP servers for specialized tool integration
39
+
> -Evaluation methods that validated agent performance
@@ -14,11 +14,11 @@ In this quest, you will set up and run a **complete end-to-end Burger Ordering A
14
14
>
15
15
> **Hackathon Award Category: Agentic System Architecture Award**
16
16
>
17
-
> As part of the Build-a-thon Hack!, we have a special award category recognizing agentic designs that are **useful, reliable, and secure**.
17
+
> This quest mapped to our special award category recognizing agentic designs that are **useful, reliable, and secure**.
18
18
>
19
-
> This quest demonstrates foundational patterns: **MCP tool integration**enables agents to interact with external systems (useful), **structured agent workflows** with proper error handling ensure consistent behavior (reliable), and **API-based architectures**allow for controlled access and validation (secure).
19
+
> The quest demonstrated foundational patterns: **MCP tool integration**enabling agents to interact with external systems (useful), **structured agent workflows** with proper error handling ensuring consistent behavior (reliable), and **API-based architectures**allowing for controlled access and validation (secure).
20
20
>
21
-
> Consider how these patterns can help you build production-ready agent systems that users can trust.
21
+
> Winners leveraged these patterns to build production-ready agent systems that users could trust.
Copy file name to clipboardExpand all lines: README.md
+20-22Lines changed: 20 additions & 22 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -2,11 +2,11 @@
2
2
3
3
<divalign="center">
4
4
5
-
5
+
6
6
7
7
<palign="center">
8
8
<strong>Create, compete, and turn ideas into working AI.</strong><br/>
9
-
<em>Starting March 2, 2026</em>
9
+
<em>March 2 – March 31, 2026</em>
10
10
</p>
11
11
12
12
<palign="center">
@@ -25,24 +25,24 @@ The JavaScript AI Build-a-thon is a focused, hands-on initiative that helps **bu
25
25
26
26
If you're building your first AI app or architecting multi-agent systems, the Build-a-thon gives you the tools, the mentorship, and the stage to ship something real.
27
27
28
-
# Key Dates
28
+
# Program Timeline
29
29
30
30
| Milestone | Date |
31
31
|---|---|
32
-
| 🎬 Build-a-thon Begins|**Monday, March 2, 2026**|
|**🔥 Hackathon Launches**|**Friday, March 13, 2026**|
35
-
|🏁 Hackathon Ends|**Tuesday, March 31, 2026**|
34
+
|**🔥 Hackathon Phase**|**Friday, March 13 – Tuesday, March 31, 2026**|
35
+
|🏆 Program Concluded|**Tuesday, March 31, 2026**|
36
36
37
-
# How It Works
37
+
# How It Worked
38
38
39
-
The Build-a-thon runs in two phases.
39
+
The Build-a-thon ran in two phases.
40
40
41
-
41
+
42
42
43
43
## 📚 Phase 1: Learn & Skill Up (March 2 – March 13)
44
44
45
-
Complete guided quests and attend expert-led livestreams to build the skills you'll need. Each quest is hands-on, self-paced, and designed to teach you a core AI development pattern with JavaScript/TypeScript:
45
+
Participants completed guided quests and attended expert-led livestreams to build the skills needed for the hackathon phase. Each quest was hands-on, self-paced, and designed to teach a core AI development pattern with JavaScript/TypeScript:
46
46
47
47
| Quest | Topic | What You'll Learn |
48
48
|---|---|---|
@@ -52,26 +52,24 @@ Complete guided quests and attend expert-led livestreams to build the skills you
52
52
|[Quest 4](/04-Build-Agents-with-AIToolKit/README.md)| Build Agents with AI Toolkit | Create AI agents with MCP tools, evaluation, and code export on VS Code |
53
53
|[Quest 5](/05-Run-Burger-Ordering-Agent-System/README.md)| E2E Burger Ordering Agent | Design and run a full agentic system with MCP server integration |
54
54
55
-
Participation Requirement: Complete at least 2 quests (including the guided project) to be eligible for the hack.
55
+
Participation Requirement: Participants were required to complete at least 2 quests (including the guided project) to be eligible for the hackathon phase.
56
56
57
57
## 🔥 Phase 2: Hack! (March 13 – March 31)
58
58
59
-
Build something that matters. Use everything you learned in the quests, and beyond, to build AI Agents that solve real problems, delight users, and push what's possible.
60
-
61
-
[Register for the hackathon](https://aka.ms/JSBuildathon_Hack)
59
+
Participants built solutions that matter using everything they learned in the quests and beyond, creating AI Agents that solved real problems, delighted users, and pushed what's possible.
62
60
63
61
64
62
65
-
During the hack phase, attend technical demos and special Discord Office Hours (8:00 AM PDT) to learn the latest tools and best practices.
63
+
During the hack phase, participants attended technical demos and special Discord Office Hours (8:00 AM PDT) to learn the latest tools and best practices.
66
64
67
-
| Date (8:00 AM PDT) | Demo |Link|
65
+
| Date (8:00 AM PDT) | Demo |Archive|
68
66
|---|---|---|
69
-
| March 16 | Azure & Foundry MCP Server Demo |[Join event](https://discord.gg/microsoftfoundry?event=1476309470312136704)|
0 commit comments