File size: 13,343 Bytes
6bff5d9
 
 
 
 
 
 
35d3523
 
 
 
 
 
 
 
 
 
6bff5d9
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
# Architecture β€” Data Eyond Agentic Service

**Last updated**: 2026-05-07
**Status**: Design phase β€” folder skeleton in place, implementation in progress

---

## Product vision (north star)

Data Eyond is an *AI data scientist* for business analytics, structured around **CRISP-DM** (Business Understanding β†’ Data Understanding β†’ Data Preparation β†’ Modeling β†’ Evaluation β†’ Deployment). Targets executives doing self-serve deep-dives and data analysts/scientists offloading routine work.

Envisioned user flow: **interview agent** captures goal β†’ user connects data sources β†’ asks natural-language question β†’ CRISP-DM-structured analytical response, exportable as a **presentation** or **notebook-style report**.

The catalog-driven, IR-based architecture documented below is the *foundation*. The next architectural evolution is an agentic layer (analytical planner, per-stage CRISP-DM agents, evaluator, reporter) that consumes the existing IntentRouter β†’ QueryPlanner β†’ Executor β†’ ChatbotAgent spine as its tool layer. See `REPO_CONTEXT.md` β†’ *Roadmap β€” agentic evolution* for the target agent topology.

---

## TL;DR

A catalog-driven AI service for data analysis. Users upload documents and register databases or tabular files; they ask natural-language questions and get answers grounded in their data.

The architecture has two paths:

- **Unstructured** (PDF, DOCX, TXT) β€” dense similarity over prose chunks (the right primitive for free-form text).
- **Structured** (databases, XLSX, CSV, Parquet) β€” a per-user **data catalog** describes what tables/columns exist; an LLM produces a structured **JSON intermediate representation (IR)** of the user's intent; a deterministic compiler turns the IR into SQL or pandas operations.

The LLM produces *intent*, not query syntax. Deterministic code does the rest.

---

## 1. Why catalog-driven design

For a database or spreadsheet, a user's question maps to *known tables and columns* β€” not to *similar text fragments*. Treating structured data with the same retrieval primitive as prose (chunk + embed + rank top-K) makes the right column survive a probabilistic ranking lottery. Catalog-based **lookup** is the right primitive instead.

A central per-user catalog also means:

- One place to keep table/column descriptions (AI-generated, refreshed when the source changes).
- The query planner sees the user's full data landscape in a single prompt.
- Schema stays stable across user sessions without hitting the source DB on every query.
- New sources auto-update the catalog without re-embedding chunks.

---

## 2. Source taxonomy

```
Sources
β”œβ”€β”€ Unstructured (pdf, docx, txt)        β†’  Cu  (prose chunks via DocumentRetriever)
└── Structured
    β”œβ”€β”€ Schema (DB)                       β†’  Cs  (DB tables + columns)
    └── Tabular (xlsx, csv, parquet)      β†’  Ct  (sheets + columns)
                                           Cs βˆͺ Ct = Data Catalog Context
```

- **Cu** = unstructured prose context. Retrieval primitive: dense similarity over chunks.
- **Cs** = DB schema context (tables, columns, descriptions, sample values).
- **Ct** = tabular file context (sheets, columns, descriptions, sample values).
- **Data Catalog Context** = `Cs βˆͺ Ct`. Passed to the query planner as a single unified view.

DB vs tabular is **not** a routing concern β€” it's a per-source attribute (`source_type`) on each catalog entry. The split only matters at execution time (SQL vs pandas).

---

## 3. Routing model

```
source_hint ∈ { chat, unstructured, structured }
```

- `chat` β€” no search, conversational reply only
- `unstructured` β€” DocumentRetriever path (Cu)
- `structured` β€” catalog-driven path (Cs βˆͺ Ct β†’ planner β†’ compiler β†’ executor)

The router commits to one path. Cross-source questions ("compare DB sales vs uploaded customer file") are handled inside the structured path because the planner sees both Cs and Ct in one prompt.

---

## 4. Core architectural decisions

### 4.1 Catalog as primary context, not retrieval

For most users (≀50 tables), the entire catalog fits in ~3-5k tokens and is passed verbatim to the planner. No vector search, no BM25, no chunk retrieval. The LLM reads the whole catalog and picks the right table.

When a user has hundreds of tables, **catalog-level retrieval** (BM25 + table-level vectors with RRF) can be added as a slicer between `CatalogReader` and `Planner`. Deferred until measurably needed.

### 4.2 JSON IR over raw SQL

The planner LLM emits a structured JSON IR describing query intent β€” not a SQL string. A deterministic compiler turns the IR into SQL (per dialect) or pandas/polars operations.

Benefits:

- Validatable with Pydantic before execution
- Compiler whitelists allowed operations (no DROP, DELETE, etc.)
- Portable: same IR β†’ SQL (any dialect) / pandas / polars
- Cheaper tokens, easier to debug, trivially testable without an LLM
- LLM cannot emit valid-but-wrong SQL syntax

### 4.3 Deterministic compiler, not LLM SQL writer

The LLM produces *intent* (the IR). All actual query construction is deterministic Python. Compiler bugs are reproducible and fixable. Same IR always produces the same query.

### 4.4 Pipeline stage isolation

Each stage is its own module with typed input and typed output. No god classes. Stages: `IntentRouter`, `CatalogReader`, `QueryPlanner`, `IRValidator`, `QueryCompiler`, `QueryExecutor`, `ChatbotAgent`. Each is testable in isolation.

### 4.5 Minimal LLM surface

LLM calls happen in exactly three places (KM-557 removed `CatalogEnricher`; ingestion is now LLM-free β€” the planner reads column names, stats, and sample rows directly):

1. **`IntentRouter`** β€” once per user message
2. **`QueryPlanner`** β€” once per structured query (produces the IR)
3. **`ChatbotAgent`** β€” once per answer (formats the response)

Compiler and executors are pure code. No LLM in the hot path of query construction.

---

## 5. End-to-end flow

### Ingestion (when user uploads a file or connects a DB)

```
source upload / DB connect
    ↓
introspect schema (DB: information_schema; tabular: file headers + sample rows)
    ↓
validate (Pydantic)
    ↓
write to catalog store (Postgres jsonb in `data_catalog`, keyed by user_id)
```

For unstructured files: chunk + embed β†’ PGVector.

### Query (per user message)

```
User message
    ↓
Chat cache check (Redis, 24h TTL)
    ↓ miss
Load chat history
    ↓
IntentRouter LLM   β†’  needs_search?  source_hint?
    ↓
    β”œβ”€β”€ chat        β†’ ChatbotAgent β†’ SSE stream
    β”œβ”€β”€ unstructured β†’ DocumentRetriever β†’ answerer
    └── structured  β†’
            CatalogReader (load full Cs βˆͺ Ct for user)
                ↓
            QueryPlanner LLM  β†’  JSON IR
                ↓
            IRValidator  (Pydantic + columns-exist + ops whitelist)
                ↓
            QueryCompiler  β†’  SQL (schema source) or pandas (tabular source)
                ↓
            QueryExecutor  (DbExecutor or TabularExecutor)
                ↓
            QueryResult
                ↓
            ChatbotAgent β†’ SSE stream
```

---

## 6. Data catalog

### Storage

Per-user JSON document, stored as a `jsonb` row in Postgres keyed by `user_id`.

### Schema (initial scope)

```
Catalog
β”œβ”€β”€ user_id, schema_version, generated_at
└── sources[]
    └── Source
        β”œβ”€β”€ source_id, source_type, name, description, location_ref, updated_at
        └── tables[]
            └── Table
                β”œβ”€β”€ table_id, name, description, row_count
                └── columns[]
                    └── Column
                        β”œβ”€β”€ column_id, name, data_type, description
                        β”œβ”€β”€ nullable
                        β”œβ”€β”€ pii_flag
                        β”œβ”€β”€ sample_values[]
                        └── stats: { min, max, distinct_count } | null
```

### Best-practice fields deferred

`description_human`, `synonyms[]`, `tags[]`, `primary_key`, `foreign_keys`, `unit`, `semantic_type`, `example_questions[]`, `schema_hash`, `enrichment_status`. Add when justified by user need.

### Stable IDs

`source_id`, `table_id`, `column_id` are stable internal references. `name` fields can change (e.g. column rename in source DB) without invalidating cached IRs.

### PII handling

Columns with `pii_flag: true` have `sample_values: null` β€” real values never enter LLM prompts. Auto-detected at ingestion via name patterns + value regex.

---

## 7. JSON IR

### Schema (initial scope)

```
QueryIR
β”œβ”€β”€ ir_version          : "1.0"
β”œβ”€β”€ source_id           : str   (references catalog)
β”œβ”€β”€ table_id            : str   (references catalog)
β”œβ”€β”€ select[]            : SelectItem
β”‚   β”œβ”€β”€ { kind: "column", column_id, alias? }
β”‚   └── { kind: "agg",    fn, column_id?, alias? }
β”œβ”€β”€ filters[]           : { column_id, op, value, value_type }
β”œβ”€β”€ group_by[]          : column_id
β”œβ”€β”€ order_by[]          : { column_id | alias, dir }
└── limit               : int | null
```

### Whitelisted operators

```
Filter ops:  = != < <= > >= in not_in is_null is_not_null like between
Agg fns:     count count_distinct sum avg min max
```

### Validation rules (enforced before execution)

- `source_id` exists in catalog for this user
- `table_id` belongs to that source
- Every `column_id` exists in that table
- Every `agg.fn` and `filter.op` is whitelisted
- `value_type` consistent with column's `data_type`
- `limit` positive int, ≀ hard cap (e.g. 10000)

If any rule fails β†’ reject IR β†’ re-prompt planner with error context (max 3 retries).

### Deferred features

`having`, `offset`, boolean tree filters (OR/NOT), `distinct`, joins, window functions. Add as user demand proves the limitation.

---

## 8. Executors

Same input (validated IR), same output (`QueryResult`), different backends.

### DbExecutor (schema sources)

```
IR β†’ SqlCompiler β†’ SQL string + params
     ↓
sqlglot validation (SELECT-only, whitelist tables/columns, LIMIT enforced)
     ↓
asyncpg / pymysql in read-only transaction with timeout (30s)
     ↓
QueryResult
```

Identifiers come from catalog (verified at validation time, safe to inline as quoted identifiers). Values are always parameterized β€” never inlined as strings.

### TabularExecutor (tabular sources)

```
IR β†’ PandasCompiler β†’ operation chain
     ↓
choose strategy by file size:
  ≀ 100 MB    β†’ eager pandas
  100 MB-1 GB β†’ pyarrow with predicate pushdown
  > 1 GB      β†’ polars lazy scan
     ↓
execute in asyncio.to_thread (CPU work off the event loop)
     ↓
QueryResult
```

Initially eager pandas is sufficient. Add the others when a real file is too big.

### Shared safety guarantees

1. IR validated before reaching compiler
2. Compiler is deterministic (no LLM)
3. Identifiers from catalog (trusted)
4. Values parameterized
5. sqlglot second-line defence for SQL
6. Read-only at every layer
7. Timeouts and row caps

---

## 9. Implementation scope

### Initial PR β€” what ships first

| Item | Folder |
|---|---|
| Data catalog Pydantic models | `src/catalog/models.py` |
| Catalog ingestion (introspect β†’ enrich β†’ validate β†’ store) | `src/catalog/`, `src/pipeline/` |
| `IntentRouter` with 3-way source_hint | `src/agents/` |
| `CatalogReader` (loads full catalog) | `src/catalog/reader.py` |
| `QueryPlanner` LLM call | `src/query/planner/` |
| JSON IR Pydantic models | `src/query/ir/models.py` |
| IR validator | `src/query/ir/validator.py` |

**Output**: a validated JSON IR object. Execution lands in a follow-up PR.

### Follow-up PRs

| PR | Scope |
|---|---|
| 2 | `QueryCompiler` (IR β†’ SQL / pandas) |
| 3 | `QueryExecutor` split: `DbExecutor` + `TabularExecutor` |
| 4 | Retry / self-correction loop on execution failure |
| 5 | Eval harness (golden question→IR→result examples) |
| 6 | Auto PII tagging in catalog |
| Later | Joins in IR, schema drift detection, hybrid catalog search |

---

## 10. Open questions

| # | Question | Why it matters |
|---|---|---|
| 1 | Catalog storage: JSON file per user vs Postgres `jsonb` row? | Affects ingestion + read performance |
| 2 | Should the catalog also list unstructured files (with descriptions only)? | Gives router unified view of all user sources |
| 3 | Catalog refresh trigger: explicit "rebuild" button, on every upload, or background TTL? | Staleness vs latency tradeoff |
| 4 | Confirm joins are out of initial IR scope? | Limits what user questions can be answered |
| 5 | PII handling for sample_values: mask, synthesize, or skip? | Affects what gets sent to LLM prompts |

---

## 11. References

- `docs/flowchart.html` β€” interactive end-to-end diagram (open in browser)
- `docs/flowchart.mmd` β€” mermaid source for the diagram

---

## Glossary

- **Cu** β€” unstructured context (prose chunks)
- **Cs** β€” schema context (DB tables/columns from catalog)
- **Ct** β€” tabular context (file sheets/columns from catalog)
- **IR** β€” intermediate representation (the JSON query shape)
- **PR** β€” pull request (a unit of code change)
- **PII** β€” personally identifiable information (names, emails, etc.)
- **ABC** β€” abstract base class (Python contract for subclasses)