Roadmap
The planned direction for Fuse, framed as intent rather than commitment, with shipped work recorded in the changelog.
The roadmap states where Fuse intends to go, ranked by impact on agent efficiency. Each planned item names its primary lever: accuracy, token cost, or round-trips. Nothing here is a commitment or a delivery date. Work that has shipped is recorded in the Changelog, not on the roadmap.
This page is for stakeholders gauging direction and engineers deciding where new work fits.
The roadmap states where Fuse intends to go, ranked by impact on agent efficiency. Each planned item names its primary lever: accuracy, token cost, or round-trips, plus rough effort and value so stakeholders can compare options. Nothing here is a commitment or a delivery date. Work that has shipped is recorded in the Changelog, not on the roadmap.
This page is for stakeholders gauging direction and engineers deciding where new work fits.
Effort and value (how to read)
Effort is a t-shirt size for one engineer who already knows Fuse, including tests and docs, not a calendar commitment.
| Size | Typical span | Meaning |
|---|---|---|
| S | 2 to 5 days | Localized change, low integration risk |
| M | 1 to 2 weeks | One subsystem, harness regression expected |
| L | 2 to 4 weeks | Cross-cutting or tuning-heavy |
| XL | 1 to 3+ months | New subsystem, ML, or agent eval loop |
Value rates expected impact on agent efficiency for the pinned benchmark corpus and typical .NET agent workflows. It is not a measured forecast.
| Value | Meaning |
|---|---|
| High | Likely large move on recall, precision, tokens, or round-trips for many tasks |
| Medium | Clear win in a subset of repos or modes; smaller headline benchmark shift |
| Low | Product polish, guardrails, or measurement without direct retrieval lift |
Re-bundling a large ONNX model, full MSBuild compilation graphs, and end-to-end agent eval each add weeks beyond the table estimate.
Planned Themes
| Item | Effort | Value | Primary lever |
|---|---|---|---|
| Auto-scope tool (task + budget, picks skeleton / focus / search internally) | L to XL | High | Round-trips; accuracy when agents pick the wrong mode |
| Session-delta emission (only new material across turns) | L | High | Token cost on multi-turn sessions |
| Hybrid retrieval with learned embeddings (opt-in rerank on BM25 top-K; sideload or small bundle, not mandatory ~90 MB default) | L opt-in; +M to L if bundled again | High on query mode for large repos with weak lexical overlap | Query recall (illustrative: Newtonsoft.Json query recall ~32% today) |
| Eval harness (task-success: tokens-to-correct-answer with and without Fuse) | XL | High for prioritization; Low direct user impact until used | Confidence; steers all other items |
Scoping Recall and Precision
BM25-only query scoping today reaches about 49% mean recall and 2% mean precision at a 50,000 token budget on the pinned PR corpus; change scoping with a git base reaches 87% recall at 50% precision. The tables below target that gap without returning to a full-repo dump.
Retrieval and seeds (recall on query mode)
| Item | Effort | Value | Expected effect |
|---|---|---|---|
Query expansion from the symbol index (stems, *Validator-style patterns, path tokens) | M | Medium to High | Best ROI without ML; helps NL queries that omit type names; modest mean recall lift, larger on FluentValidation-style repos |
| Dual recall (BM25 plus embeddings, sparse retrieval, or symbol-alias merge) | L to XL | High | Broad query recall lift when second signal is semantic; cost scales with chosen backend |
| Cross-encoder or lightweight rerank on top-K | L to XL | High on hard repos | Targets AutoMapper and Newtonsoft.Json gaps where grep vocabulary beats BM25 today |
| Optional query rewrite (task description to keywords and type hints) | M | Medium | Helps prose-heavy agent prompts; LLM-backed rewrite adds policy and cost |
Primary lever: accuracy on query mode. Secondary: round-trips when one call replaces blind grep loops.
Dependency graph (recall without false expansion)
The syntax-level Roslyn graph is the ceiling for focus, changes, and query expansion. Large repos with conditional compilation and partial classes trail smaller ones in recall.
| Item | Effort | Value | Expected effect |
|---|---|---|---|
| Tighter false-edge control (less noise from stripped comments and literals) | M | Medium | Fewer wrong expansions; helps Newtonsoft.Json-style graphs |
| Edge confidence and typing (strong vs weak edges, weighted expansion) | M to L | Medium to High | Better precision at same budget; steadier recall on large corpora |
| Compiled reference resolution (MSBuild or Roslyn when a solution exists) | XL | High | Structural ceiling lift for cross-assembly and resolved types; maintenance heavy |
Primary lever: accuracy on changes and focus, especially on large corpora.
Expansion and ranking (recall vs precision trade)
Expansion protects recall by pulling neighbors; neighbors outside the ground-truth changed set hurt benchmark precision even when they help an agent.
| Item | Effort | Value | Expected effect |
|---|---|---|---|
| Tighter admission thresholds | S to M | Medium | Higher benchmark precision; risk of recall drop if over-tightened |
| Adaptive depth (deepen only high-scoring branches) | M | Medium | Better recall per token than fixed depth everywhere |
| Selective reverse hops for query (callers/tests when seed is strong, capped) | M | Medium | Recall lift for tests and callers without focus-wide broadening |
| Two-phase expansion (must-have deps first, optional neighbors if budget remains) | M to L | Medium to High | Precision at fixed budget with less recall sacrifice than blunt depth cuts |
| Learned or feature-based reranking after expansion | L to XL | Medium to High | Feature plumbing M to L; learned rank XL; rescores tail before budget cut |
Primary lever: precision at a fixed budget without sacrificing headline recall on small repos.
Emission tiers (precision inside selected files)
File selection can stay recall-oriented while emission trims noise inside each file.
| Item | Effort | Value | Expected effect |
|---|---|---|---|
| Stronger member-level packing for query | S to M | Medium | Same file recall, fewer tokens and less sibling noise in matched files |
| Path and test policy (tests only when query, seed, or change signals them) | S to M | Medium | Precision and token savings; may drop useful test context if heuristics are crude |
| Score-tiered emission (full / outline / skeleton by rank within the included set) | M to L | High | Large token reduction at same file set; improves effective precision per token |
Primary lever: token cost and effective precision (less noise per token) at unchanged file recall.
Workflow and product surfaces (agent round-trips)
| Item | Effort | Value | Expected effect |
|---|---|---|---|
| Task-shaped defaults (PR and branch work steers to change scoping when a git base exists) | S | High | No algorithm change; moves many PR tasks from ~49% query recall to ~87% changes recall on the harness |
| Guided multi-call flows (TOC or skeleton, then focus or changes on large repos) | S to M | Medium to High | Extra round-trip but better accuracy when single query at 50k is insufficient |
Primary lever: round-trips and accuracy in live agent sessions without requiring 100% single-call recall.
Measurement
| Item | Effort | Value | Expected effect |
|---|---|---|---|
| Per-repo recall dashboards in harness output | S | Low direct; High for dev | Surfaces Newtonsoft.Json and AutoMapper regressions hidden by the mean |
| Layer 2B regression guard when retrieval ships | S; M if wired to CI compare gate | Low direct; High for dev | Catches lexical localization regressions (grep baseline is 58% on 12 questions today) |
| End-to-end task success (see Planned Themes eval harness) | XL | High for prioritization | Validates that harness recall moves translate to agent task success |
Primary lever: confidence that accuracy work improves real tasks, not only harness aggregates.
Phased bundles (effort vs value)
Illustrative sequencing for one engineer. Calendar time overlaps if items run in parallel.
| Phase | Scope | Effort (cumulative) | Value | Notes |
|---|---|---|---|---|
| A Quick wins | Query expansion, admission tuning, tiered emission v1, task-shaped MCP defaults, per-repo harness reports | ~3 to 5 weeks | High value per week | No ML; best first release |
| B Expansion and graph hygiene | Adaptive depth, two-phase expand, edge weights, false-edge fixes | ~4 to 8 weeks after A | Medium to High | Large-repo focus and changes/focus more than raw query |
| C Semantic retrieval v2 | Opt-in hybrid rerank on BM25 top-K (no mandatory bundle) | ~6 to 10 weeks | High on query on hard repos | Reuses prior hybrid design lessons; packaging optional |
| D Compilation graph | MSBuild or Roslyn references where available | ~3 to 6 months | High structural ceiling | Independent track; edge cases and maintenance |
| E Learned ranking and task eval | LTR or cross-encoder plus agent eval harness | ~3 to 6+ months | High when eval proves lift | Depends on eval design |
Fastest benchmark moves: task-shaped defaults (S effort, High value, already shipped capability); query expansion plus top-K rerank (M plus L, High value on query); tiered emission (M to L, High token value). Slowest but deepest: compiled graph (XL) and full task eval (XL).
Sequencing intent (not dates)
Near-term emphasis: phase A items. Medium-term: phase B and opt-in hybrid rerank (phase C). Long-term: compilation graph (phase D), dual recall, learned ranking, and task eval (phase E).
Recently Shipped (no longer on the roadmap)
These items moved to the Changelog when they shipped:
- Roslyn structural analysis for C# (skeleton, dependency graph, type location, outlines, route maps, semantic markers) is the default path; regex remains for content reduction, the project graph, and pattern detectors.
- Symbol-level scoping (
Type.Memberfocus seeds and member-granularity query retrieval). - A persistent SQLite store at
.fuse/fuse.dbfor reduction output, analysis, and relevance tokens.
How This Relates to Shipped Work
The roadmap describes intent. Once an item ships, it is removed from the roadmap and documented in the Changelog with the version that introduced it. To learn what Fuse does today rather than what is planned, read the changelog and the reference pages.
What This Does Not Cover
This page does not specify designs or release dates beyond the phased bundles under scoping recall and precision. Effort sizes are illustrative, not estimates tied to staffing or calendar. Shipped behavior is documented in the Changelog.
Next
Read the Changelog for the history of shipped work, or Contributing to propose or build a planned item.