AI-assisted planning for business decisions
A planning tool that uses past software delivery work to estimate new requests and flag possible delays. Leaders can review changing estimates and workloads in one dashboard.
- Context
- Internal planning tool at a B2B marketplace
- My role
- Hands-on builder
- When
- 2025–2026
- re-estimation on every Jira change
- Real-time
- delivery estimates and workloads for leadership
- 1 dashboard
I built this for the leadership team of a B2B marketplace with an internal software team. It is an example of bringing existing business data together to support decisions.
The challenge
The leadership team had to decide what to build next without a reliable answer to “when will it actually be done?”. Estimates were inconsistent, and delays showed up too late to react.
What I built
- Similarity-based estimates. When a ticket is created or changed, a vector search finds similar completed tickets. Their real effort patterns are fed to an LLM, which produces an estimate.
- Estimates at every level: epics, subtasks, research and design work, with AI estimates as the fallback in the delivery timeline.
- Updates when plans change. Changes in Jira, the team’s work-tracking tool, trigger a new estimate.
- Decision dashboards for leadership: utilisation, builder performance, and AI analysis of each work package.
- AI-assisted intake. New requests are pre-filled by AI analysis, and semantic search finds related work instead of relying on exact keywords.
- Production-grade LLM plumbing: model routing with fallbacks and tracing of every LLM call.
Why it matters
The tool puts estimates and workload information in one place and updates them as work changes. Leaders still decide priorities and assess whether an estimate is realistic.