What you can build with AI today

Inputs

• Domain image–text pairs
• Base multimodal checkpoint
• Fine-tuning stack

Steps

1. Prepare paired images and captions for your domain.
2. Load the open-weight multimodal base model.
3. Run supervised fine-tuning on your dataset.
4. Evaluate on held-out prompts before deployment.

Tools

Inkling

Evidence

• Introducing Inkling documentation

AI Signal article →

Inputs

• PDF or Markdown files
• Embedding model
• Vector store
• LLM for generation

Steps

1. Chunk documents into retrievable segments.
2. Embed chunks and load them into a vector index.
3. Retrieve top matches for each user question.
4. Prompt the LLM with retrieved context and cite sources in the answer.

Tools

LlamaIndex

Evidence

• LlamaIndex data framework repository

Inputs

• Eval registry entries
• Model API or local endpoint
• Scoring rubric per eval

Steps

1. Pick or author evals in the OpenAI Evals format.
2. Configure the model and sampling settings.
3. Execute the eval run and capture completions.
4. Aggregate scores and diff results between runs.

Tools

OpenAI Evals

Evidence

• OpenAI Evals framework repository

Inputs

• Local git repository
• Task description
• Codex CLI or IDE integration

Steps

1. Open the target repo in a Codex-enabled environment.
2. Describe the change with file paths and constraints.
3. Review the proposed diffs before applying.
4. Run project tests and iterate on failures.

Tools

Codex

Evidence

• OpenAI Codex CLI repository

Inputs

• Product description or style brief
• Qwen-Image model access
• Reference constraints (optional)

Steps

1. Write a prompt describing the product shot you need.
2. Run Qwen-Image to produce candidate renders.
3. Compare variations and refine the prompt.
4. Export selected images for mockups or listings.

Tools

Qwen-Image

Evidence

• Qwen-Image model repo repository
• Qwen-Image 3.0 overview documentation

AI Signal article →

Inputs

• Source text or documents
• Pydantic (or similar) schema
• LLM with Instructor client

Steps

1. Define the output schema for the fields you need.
2. Wrap your LLM client with Instructor.
3. Prompt for the structured object in one call.
4. Validate the response against the schema and handle retries on failure.

Tools

Instructor

Evidence

• Instructor structured outputs repository

Inputs

• Workflow graph definition
• Tool or API integrations
• Checkpoint store

Steps

1. Model the workflow as nodes and edges in LangGraph.
2. Attach tools and state reducers to each node.
3. Compile the graph with persistence for checkpoints.
4. Resume or branch runs from saved state when inputs change.

Tools

LangGraph

Evidence

• LangGraph agent orchestration repository

Inputs

• Embedding vectors
• FAISS index type choice
• Query vector

Steps

1. Normalize or prepare vectors for the chosen metric.
2. Build a FAISS index for your corpus size and recall needs.
3. Add vectors to the index in batches.
4. Run top-k similarity queries and map hits back to records.

Tools

FAISS

Evidence

• FAISS vector search library repository

Inputs

• Source image
• SAM checkpoint
• Point, box, or mask prompts

Steps

1. Load the Segment Anything model and image encoder.
2. Place prompts on the target region.
3. Generate candidate masks from the prompts.
4. Pick the best mask and export cutouts or overlays.

Tools

Segment Anything

Evidence

• Segment Anything Model (SAM) repository

Inputs

• Audio file (wav, mp3, etc.)
• Whisper model checkpoint

Steps

1. Install Whisper and pick a model size for your hardware.
2. Load the audio and run transcription inference.
3. Review the transcript and segment timestamps if needed.
4. Export text for subtitles, notes, or downstream search.

Tools

Whisper

Evidence

• Whisper speech recognition repository