Free forever, no credit card.Get Started for Free →
← All posts
November 22, 2026 · 6 min read

Does Jitterbit Remember Between Runs? The Watermark Holds. The Reasoning Does Not.

Does Jitterbit Remember Between Runs? The Watermark Holds. The Reasoning Does Not. Ask whether Jitterbit remembers between runs and the answer splits cleanly in two. Jitterbit remembers where a sync stopped, in careful detail, because moving records between systems without losing or duplicating them is its entire job. It does not remember what an automation or an AI agent was trying to accomplish with those records, why it handled an exception the way it did, or what it already learned on the l

Does Jitterbit Remember Between Runs? The Watermark Holds. The Reasoning Does Not.

Ask whether Jitterbit remembers between runs and the answer splits cleanly in two. Jitterbit remembers where a sync stopped, in careful detail, because moving records between systems without losing or duplicating them is its entire job. It does not remember what an automation or an AI agent was trying to accomplish with those records, why it handled an exception the way it did, or what it already learned on the last run. Those are different kinds of memory, and confusing them is how scheduled integrations re-decide the same edge cases every cycle.

What Jitterbit actually keeps

The clearest piece of Jitterbit memory is the high-watermark. In the documented incremental sync pattern, an operation stores a single value, such as the maximum LastModifiedDate from the previous batch, and reads it back at the start of the next run. Two cache functions do the work: ReadCache retrieves the stored watermark, and WriteCache updates it after processing completes. Only new or modified records are retrieved on the next run, so a Salesforce-to-ERP sync touches what changed overnight instead of re-reading the entire account. The cache expiration in the pattern is 30 days, which is also the maximum, and the scope decides who can see the value: project scope makes it accessible to all operations in the project and persists it between separate runs.

A second pattern uses the same cache for change detection. A hash of each record, computed with SHA256 or MD5, is stored from one run and compared on the next. If the hash matches, the record has not changed and can be skipped. If it differs, the record is processed and the new hash is written back. The hash is a fingerprint, nothing more. It proves a record changed. It says nothing about why it changed, whether the change matters, or what anyone decided about it last time.

Jitterbit also keeps a record that runs happened. Every operation execution generates an operation log with when and where the operation ran, its status, the number of source and target records processed, and any log messages. Harmony retains operation logs for 30 days, and the log view shows a maximum of 1,000 logs for a given timeframe. Audit logs follow the same shape: user activity across the Management Console, API Manager, and Studio is retained for 30 days. All of this is an operations record. It answers what ran, when it ran, and how it ended.

Even conversation history inside Jitterbit is explicit manual work. The documented pattern for a multi-turn LLM chat stores the full exchange in Cloud Datastore as rows in a CSV string, keyed by a session identifier, then retrieves and reassembles that history before each call and writes it back afterward. A native LLM connector can retain chat context across operations that share the same chatId, and private agents retain that context in memory. In every case the boundary is the same: history survives only where someone deliberately stored it, inside that one design.

Where the watermark stops

The watermark is exact about data movement and silent about everything around the movement. Consider a scheduled order sync with an AI step that reviews exceptions: records that fail validation, customers whose details do not match across systems, items that need a human decision. Jitterbit tells the next run which records changed since the stored watermark. It does not tell it which exception yesterday's run already investigated, what a support conversation established about that customer, which mismatch a person already ruled harmless, or why a particular order was routed one way instead of another. The run reads the changed records perfectly and still re-investigates the same exception from zero, because the reasoning that resolved it never became part of any record the run can read.

The cache limits sharpen the point. A watermark or a hash that expires after 30 days is a short horizon for a long-running operation. A monthly sync that slips, a paused project, or a seasonal workflow can return to find its stored position gone, and the safe response is to reprocess broadly and deduplicate. That is sound engineering for data movement. It is not a memory of the work.

The gap widens when several tools share the work. A Jitterbit operation, an agent in n8n watching the same accounts, and a reporting job in another tool can all touch the same order and still hold three different understandings of it. Each system keeps its own log, its own cache, or nothing at all. The watermark is shared with no one outside its scope. The reasoning is shared with no one at all, and every run that starts cold has to reconstruct it from raw records, invent it, or skip it and act on the data alone.

What a shared memory layer adds

The fix is not to replace the watermark. The watermark does its job well. The fix is to give the runs themselves a memory that sits alongside Jitterbit, in a place every tool in the workflow can reach.

That is the role of Vilix AI. Vilix AI is a cloud-hosted memory layer for AI agents and automations, reached over MCP, so there is no database to provision, no vector store to maintain, and no infrastructure to run. A Jitterbit operation's AI step, an n8n workflow, and a chat assistant can all connect to the same Vilix AI memory and read the same shared state, instead of each keeping a private cache the others cannot see.

What gets stored there is the kind of memory Jitterbit does not keep: the full conversation history and working context, not just extracted facts or fingerprints. When an exception gets investigated, the investigation itself is saved. When a run decides how to handle a mismatched customer, the decision and its reasons are saved. The next run, in any connected tool, retrieves what is relevant to the records in front of it and starts from what is already known instead of from a watermark and a blank page.

Getting started takes minutes. Create an account, connect Vilix AI as an MCP server in the tools that run the automations, and let each run save its conclusions as it works. The free plan is free forever, and the Pro plan comes with a 7-day trial with no card required, so the setup can be tested against a real scheduled operation before any commitment. Memory stays under the operator's control: it can be exported or deleted at any time.

The practical test

A useful test for any automation stack is simple. Pick the last exception an operation handled more than once and ask where the knowledge from the first handling lives. If the answer is a cache value, a hash, or a log entry with a count and a timestamp, the stack remembers that work happened. It does not remember the work. Jitterbit will keep holding the watermark, exactly and reliably, for as long as the operations run. Give the runs a memory of their own in Vilix AI, and the next run stops re-asking questions the last run already answered.

Get Started for Free

Persistent memory across ChatGPT, Claude, and the AI tools you already use in Vilix AI.

Get Started for Free

Free forever, no credit card.

Keep reading
Does Informatica Remember Between Runs? The Parameter Holds. The Reasoning Does Not.

Does Informatica Remember Between Runs? The Parameter Holds. The Reasoning Does Not. Ask whether Informatica remembers between runs and the answer splits cleanly in two. Informatica remembers where a data load stopped, in precise and deliberate detail, because incremental integration without losing or duplicating rows is its entire job. It does not remember what an automation or an AI agent was trying to accomplish around that load, why it handled an exception the way it did, or what it already

Does Zuora Remember Between Runs? The Subscription Record Holds. The Run Reasoning Does Not.

Does Zuora Remember Between Runs? The Subscription Record Holds. The Run Reasoning Does Not. Ask whether Zuora remembers between runs and the answer splits cleanly in two. Zuora remembers the subscription, in versioned detail, because billing correctly across months and amendments is its entire job. It does not remember what an automation or an AI agent was trying to accomplish around that subscription, why it handled an exception the way it did, or what it already learned on the last run. Thos

Does Celigo Remember Between Runs? The Watermark Holds. The Reasoning Does Not.

Does Celigo Remember Between Runs? The Watermark Holds. The Reasoning Does Not. Ask whether Celigo remembers between runs and the honest answer splits in two. Celigo remembers where a flow stopped, in precise detail, because moving records between systems without losing or duplicating them is its entire job. It does not remember what an automation or an AI agent was trying to accomplish with those records, why it handled an exception the way it did, or what it already learned yesterday. Those a