Workflow · saas · workflow

Building Your Own Product Copilot: Challenges, Opportunities, and Needs

Software engineers building product copilots face pain points at every stage: prompt engineering is fragile and time-consuming, orchestrating multi-step AI workflows is hard to steer, and existing software engineering processes have not caught up with the demands of building AI-powered applications.

How it works
Common implementation structure
How this type of workflow is generally built, generalized across documented cases — not tied to any one vendor's stack. Click any stage to read what happens there. Specific products that implement these stages appear in “Tools commonly seen” below.
Stage 1 · User natural-language query
The user provides a query or command, and the copilot provides a response in a single-turn interaction.
Tools used
GPT-4ClaudeLLaMAGitHub CopilotMural
Outcome

The study produced collaborative findings and tool design directions for the software engineering community, with participants converging on prompt asset management, orchestration tooling, and AI-specific benchmarking as the highest-priority unmet needs.

What failed first

Models produce inconsistent, hallucinated, or malformed outputs; get stuck in loops; and mistakenly signal task completion prematurely, while unit-testing frameworks designed for deterministic code cannot cope with generative model variability.

Results
Volume10
Cost replaced1-2 cents
Source

https://arxiv.org/html/2312.14231

How we source this →

Grounding & classification
Source type: technical build writeup
22 fields verified against source quotes.
agentic workflowcode generationconversational aicode diff prfailure mode describedhuman review describedmetric backedpeer confirmedproduction runtime claimedtools describedworkflow describedsoftwaretechnical build writeupagentic task executionextract classify route