Showcases

Explore 27 executable workflows organized by technical domain. Each page explains the workflow DAG, activity responsibilities, inputs, outputs, provenance, and execution considerations without embedding implementation code in the documentation.

The complete Dockerfiles, Python programs, generators, input fixtures, runners, and validators are maintained separately in UFFeScience/akoflow-examples. A single source-bundle link on each page connects the explanation to its versioned implementation.

Choose a category

• Machine Learning — training, tuning, inference, evaluation, conversion, vision, ensembles, and infrastructure selection.
• Generative AI — retrieval, synthetic data, audio transcription, document extraction, and offline video creation.
• Document AI — classification, structured extraction, and reconciliation.
• Media intelligence — reproducible video inspection, segmentation, audio analysis, and timelines.
• Rendering — procedural scenes, parallel frame shards, and animation composition.
• Agentic workflows — bounded agents, supervised tool use, and multi-agent review.
• Scientific AI — AI-assisted analysis, surrogate models, and experiment feedback loops.

Machine Learning

Executable training, tuning, inference, evaluation, and model-delivery workflows.

ingest data↓validate↓train↓evaluate↓

register modellog MLflow run

Training with MLflow

↗

prepare dataset↓

trial 1trial 2trial N

↓compare trials↓register best

Hyperparameter search

↗

fetch base model↓prepare examples↓fine-tune adapters↓evaluate↓

publish adapterrecord lineage

LoRA fine-tuning

↗

load model↓partition input↓

infer shard Ainfer shard Binfer shard N

↓merge predictions↓publish

Batch inference

↗

freeze evaluation set↓

evaluate Aevaluate Bevaluate C

↓compare metrics↓approve candidate

Comparative model evaluation

↗

load model↓

export ONNXexport TorchScript

↓

quantize CPUquantize GPU

↓validate outputs↓publish variants

Model conversion and quantization

↗

collect images↓import labels↓augment↓train detector↓evaluate↓package model

Computer vision training

↗

collect signals↓build features↓train detector↓score validation↓choose threshold↓publish

Anomaly detection

↗

prepare input↓

predict Apredict Bpredict C

↓combine↓evaluate↓publish

Model ensemble

↗

load teacher↓soft labels↓train student↓

qualitylatency

↓publish

Model distillation

↗

freeze inputs↓

CPUGPUedge

↓normalize↓compare cost and latency

Inference benchmark

↗

profile workload↓

local GPUHPCcloud

↓compare plans↓select scope↓train

Infrastructure selection for training

↗

Generative AI

Executable data preparation and evaluation workflows for retrieval, generation, audio, and documents.

collect documents↓clean↓chunk↓embed↓build index↓

retrieval testsanswer tests

↓publish

RAG index construction and evaluation

↗

define schema↓

batch Abatch Bbatch N

↓filter↓deduplicate↓quality review↓publish

Synthetic dataset generation

↗

ingest audio↓segment↓

segment 1segment N

↓merge transcript↓summarize↓publish

Audio transcription and summarization

↗

ingest documents↓

OCRnative text

↓classify↓extract fields↓validate schema↓publish

OCR and document extraction

↗

storyboard↓

scenesnarration

↓compose↓quality control

AI Video Studio

↗

ingest↓chunk↓index↓retrieve↓answer↓evaluate

Local RAG Factory

↗

Document AI

Executable extraction and reconciliation workflows over auditable document corpora.

ingest↓classify↓extract↓reconcile↓report

Document Intelligence

↗

Media intelligence

Offline audiovisual inspection, segmentation, measurement, and timeline workflows.

prepare↓

probescenesaudio

↓timeline↓report

Video Intelligence

↗

Rendering

Procedural scene construction and parallel frame rendering workflows.

setup↓

even framesodd frames

↓compose↓verify

Blender Render Farm

↗

Agentic workflows

Executable bounded-agent workflows with explicit review, policy, and evidence stages.

prepare objective↓agent loop↓validate output↓

publish resultstore trace

Encapsulated agent task

↗

build context↓propose action↓policy check↓human approval↓execute tool↓verify

Supervised tool loop

↗

shared context↓

researchcodedomain

↓critic review↓synthesize↓publish evidence

Multi-agent review

↗

Scientific AI

Executable simulation, analysis, surrogate-model, and experiment-feedback workflows.

prepare experiment↓simulation↓measurements↓

statisticsAI interpretation

↓validate↓report

Scientific pipeline with AI interpretation

↗

design samples↓

simulation Asimulation N

↓assemble dataset↓train surrogate↓validate↓publish

Surrogate model construction

↗

observations↓update model↓propose experiments↓safety check↓run experiments↓record observations

Model-guided experiment cycle

↗

Validation model

Every source bundle is exercised by the centralized examples CI. The regression fixture verifies its declared local inputs and outputs; provider-specific execution on HPC, Kubernetes, or cloud infrastructure still depends on the environment and permissions described by the corresponding infrastructure guide.