Winner · Grand Challenge Award

HAIEC wins
MunichTech
Grand Challenge Award

AI Action Path Assurance recognized as the overall best project at Europe's premier applied AI innovation platform.

MUNICHTECH EXPO WINNER GRAND CHALLENGE AUTUMN 2026
Dateline Munich, Germany
Date 26 September 2026
Award Grand Challenge
Score 8.7 / 10
The Platform · MunichTech EXPO

Europe's applied AI and deep-tech innovation platform, where proof-of-concept meets production.

Reach
5,000+
Participants across Europe, the Gulf, and beyond
Exhibitors
300+
Startups, scale-ups, and enterprise teams
Audience
Industry
Corporates, investors, governments, and researchers
Focus
Production
From prototype to real-world deployment
Ecosystem & Partners
OpenAI AI Platform
ElevenLabs Voice AI
Featherless Infrastructure
RISE Research Applied Research
SRTIP UAE Innovation
BMW Group Industry Partner
01 · The Story

AI systems no longer answer. They act.

Context MunichTech EXPO
Autumn 2026 Hackathon
20-day build sprint

Modern AI agents don't just reply. They call tools. Invoke APIs. Modify databases. Trigger workflows. Change real system state.

Yet most assurance tools still look at fragments. The model, the permission, the scanner finding. They miss the harder question.

What can this AI-enabled system actually reach and cause?

HAIEC was built to answer that question. Not with a diagram. Not with a synthetic example. With a working platform validated against a real multi-tenant AI SaaS application.

"
We did not want to demonstrate this with a diagram or a synthetic example. We built something that works on real systems. Something enterprises can actually inspect, verify, and trust.
Subodh KC · Founder, HAIEC
Consequence Path From intent to
real-world effect

The full path, made visible.

01
Intent · Stage 01
Human or business goal
What the system was asked to do, and by whom. The starting point of every action path.
02
Intent · Stage 02
AI system or agent
The model, orchestration, or reasoning loop that interprets intent and decides on action.
03
Mechanism · Stage 03
Tool, API, or MCP server
The interface boundary. Where the AI system reaches out to touch the world.
04
Mechanism · Stage 04
Handler and implementation
The application code that executes the call. Where intent becomes machine action.
05
Mechanism · Stage 05
Policy and application controls
The guardrails, feature flags, and policy checks that sit between intent and effect.
06
Effect · Stage 06
Identity and granted authority
The credential, role, and scope that actually gives the action permission to execute.
07
Effect · Stage 07
Real-world consequence
The observable effect on systems, data, money, or people. The point HAIEC exists to make visible.
02 · Assurance Model

Five independent planes, never collapsed.

Capability is not permission. Permission is not execution. HAIEC keeps each claim separate, advancing each only as far as the evidence supports.

01
Requested
Was this action actually declared or requested?
02
Policy Authorized
Was this capability approved for this use and scope?
03
Effectively Granted
Did the acting identity actually have the authority?
04
Code Capable
Can the evaluated implementation reach the consequence?
05
Observed
Do qualifying logs, traces, or witnesses show it actually happened?
03 · Validation

Tested against a real multi-tenant AI SaaS.

HAIEC was evaluated against KestrelVoice, a working multi-tenant AI SaaS application, using a frozen source snapshot for reproducibility.

44/44
Action paths met criteria
1,322
Supporting evidence traces
555
Verified checkpoints
32
Deterministic checks
99
Unresolved facts preserved
8.7/10
Jury score
04 · Scoring

Judged across five criteria.

Evaluated by an independent jury under the published Devpost Grand Challenge framework.

Problem Relevance & Impact
9.0
Technical Excellence & Feasibility
9.0
Innovation & Originality
8.5
Practical Applicability & Scalability
8.5
Presentation & Communication
8.5
Overall average
8.7
05 · The Team

Built by a small, focused team.

SK
Subodh KC
Founder · Lead
KG
Kushal Gautam
Engineering
RB
Roshan Basnet
Engineering
AY
Abhishek Yadav
Engineering
06 · Recognition

Award package.

01
Official award recognition at MunichTech EXPO 2026
02
On-stage award ceremony and public announcement
03
Featured project showcase at the EXPO
04
Visibility across MunichTech EXPO website, press, and social channels
05
Introductions to industry partners, enterprises, investors, and public-sector stakeholders
06
Fast-track opportunities for exhibiting, pitching, or joining the MunichTech EXPO Talent Hub
For Enterprise Buyers

See what your AI can actually cause.

HAIEC is the exposure control layer for enterprises deploying consequential AI systems. Request a demo to see the evidence-backed path from AI capability to real-world consequence.

Request a Demo View the Project
About HAIEC Human AI Evidence Company

HAIEC builds deterministic assurance for AI systems by connecting evaluated scope, evidence, capability, assurance claims, and verifiable decision records.

Unlike checklist tools or black-box scoring, HAIEC generates reproducible evidence mapped to real regulations and enterprise governance requirements.

Founded by Subodh KC, a former Fortune 100 AI strategy leader, HAIEC was built from real-world experience implementing AI compliance at enterprise scale, not from theory.

HAIEC
Human AI Evidence Company
MunichTech EXPO Grand Challenge Winner · Autumn 2026
Press contact: subodhkc@subodhkc.com
haiec.com Devpost project Award announcement Newsroom Project gallery Hackathon Print PDF