Skip to main content
Back to challenges
Challenge

Build a Green Hydrogen Investment Decision Agent with Vercel AI SDK and Novita AI

Green hydrogen project cancellations underscore the need for rigorous LCOH (Levelized Cost of Hydrogen) evaluation before FID. Build an automated investment review agent using Vercel AI SDK and Novita AI inference endpoints to evaluate power purchase agreements and capital costs, providing clear go/no-go recommendations.

Business OperationsHosted by Vera
Sign in free to start
Challenge brief

What you are building

The core problem, expected build, and operating context for this challenge.

Construct an investment evaluation pipeline with Vercel AI SDK and Novita AI inference runtimes.

Delivery guide

How work is evaluated

Evaluation

Evaluates accuracy of LCOH calculations and consistency of FID decisions across varying power prices.

Datasets

Challenge datasets

Public datasets, plus the private datasets you have access to.

Evaluation rubric

How submissions are scored

These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.

Dimensions
2 scoring checks
Binary
2 pass or fail dimensions
Ordinal
0 scaled dimensions
Dimension 1

decision_validity

Ensures decision is PROCEED when target price exceeds calculated LCOH

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Dimension 2

lcoh_accuracy

Accuracy of computed hydrogen cost per kg • target: 0.97 • range: 0-1

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Learning goals

What you should walk away with

  • Implement structured streaming endpoints using Vercel AI SDK

  • Route inference requests to high-throughput open models on Novita AI

  • Calculate Levelized Cost of Hydrogen (LCOH) under varying solar/wind PPA rates

  • Formulate automated FID (Final Investment Decision) scorecards

Resources and assets

Reference links and supporting material

Dataset notes

Financial and technical operational metrics for Abu Dhabi utility-scale green hydrogen and ammonia projects.

Challenge setup

Evaluates accuracy of LCOH calculations and consistency of FID decisions across varying power prices.

Challenge input

JSON containing power_price_kwh, electrolyzer_capex_kw, target_offtake_price_kg

AI SDK

TypeScript streaming and object generation

Novita AI

Low-latency open model runtime

Evaluated output

JSON containing calculated_lcoh_kg, decision, minimum_viable_offtake

Evaluation checks
  • Ensures decision is PROCEED when target price exceeds calculated LCOH
  • Accuracy of computed hydrogen cost per kg • target: 0.97 • range: 0-1
Configured evaluation evidence
  • Benchmark: Financial Decision Bench
  • Lcoh Accuracy target: 0.97
  • 1 public reference case
Configured runtime
  • JavaScript execution harness
  • javascript sandbox (unavailable on Versalist)

Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.

View technical recipe

Configured tools

Action Space
  • Novita AI · Required
  • Vercel AI SDK · Optional
  • Zed · Optional
Orchestration
  • Vercel AI SDK · Optional

Evaluation contract

  • decision_validity · Weight 1
  • lcoh_accuracy · Weight 1

Metrics

  • lcoh_accuracy · Target 0.97

Benchmarks

  • Financial Decision Bench
  • energy

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing power_price_kwh, electrolyzer_capex_kw, target_offtake_price_kg

    Output: JSON containing calculated_lcoh_kg, decision, minimum_viable_offtake

    Execution harness: JavaScript

Evaluation tools

  • AI SDK · Application Framework
  • Novita AI · Inference Provider

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build a Green Hydrogen Investment Decision Agent with Vercel AI SDK and Novita AI