Agent stuck on a task for 24 hours
The mentor spots a task that hasn't progressed since yesterday, asks the agent what's blocking, and either resolves it or surfaces a clear next step to you.
Sistava Mentor
Every employee you hire on Sistava works alongside Steve, the Sistava Mentor: a named, Sista-side coach who checks in on their work and posts straight into the same chat thread you already read. When a hire drifts from the brief, misses a step, or gets stuck, Steve sends a message under his own name and avatar coaching them on craft, never on priorities. You always see it happen: the message sits in your chat history like any other turn, so nothing about the correction is hidden from you.
The mentor is a Sistava staff member assigned to your workforce from day one. You do not hire, pay for, or manage them. They work alongside your team to provide continuous guidance and support.
They regularly check each employee's progress, share important updates, answer questions, and identify anything preventing work from moving forward. When an employee needs help, the mentor responds and gets them back on track.
They also handle technical issues, recurring failures, concerns, and complaints as soon as they appear. Your team leader continues managing performance, while the mentor ensures every employee remains supported and productive.
Pairs with every employee, automatically
The moment you hire a new AI employee, the mentor is paired with them. No setup, no configuration, no extra cost. Open the employee's profile and you'll see him listed right next to who they report to. The pairing is universal: the same mentor for everyone on your roster, so there's a consistent standard of care across the workforce.
He stays paired for the lifetime of the employee through new hires, transfers between teams, and role changes. The mentor follows along, building context on each agent's work over time so his check-ins get more accurate the longer they're together.
Periodic check-ins that catch blockers early
The mentor runs a routine sweep across every employee on your roster. He looks at the work board, the recent conversation, the activity feed, the tools the agent has been using. He's looking for the patterns that mean something is wrong: tasks frozen mid-flight, retries piling up, the same error appearing across days, the agent waiting on an input that never came.
When he finds one, he doesn't just log it. He nudges the agent, asks what's blocking it, and tries to resolve it on the spot. If he can't, he surfaces it to you with context: a specific problem with a recommended next step, not a vague alert. You spend less time monitoring and more time on actual work.
The fallback when the team lead is busy
Team leaders coordinate work, not micromanage it. When an agent has a question, such as a clarification on what to do next, a judgment call on tone, or a sanity check on a draft, they normally route it to their leader. But leaders run sprints, manage multiple direct reports, and aren't always there to answer in the moment.
The mentor is the second escalation. If the leader is busy or the question is more about how to work than what to work on, the agent talks to the mentor instead. Work keeps moving. The leader gets a summary later if it's relevant, and you don't get pinged for things that don't need you.
A quality layer that complements guardrails
Approval gates, content policies, and tool limits are mechanical guardrails. They block specific actions that match specific rules. The mentor is the human-shaped layer above them. He's looking at the things rules can't see: is the agent's tone consistent with how you set it up, is the work actually getting toward the goal, is the conversation going in circles, or did the agent forget something important from a previous turn.
Rules catch the obviously wrong. The mentor catches the subtly off. Together they make a workforce that doesn't just stay safe, it stays on track.
How It Works
Every AI employee is paired with a mentor who sweeps the workforce on a schedule, catches blockers, answers the agent's questions when the leader isn't around, and surfaces what needs your attention.
The mentor runs as a scheduled background routine on the platform. On each sweep he reads the work board, the activity feed, and the recent conversation for every employee on your roster. He looks for tasks that haven't moved, questions that went unanswered, and tools that keep failing. He nudges the agent directly when he can resolve something on the spot, and surfaces the rest to you with context.
When an employee has a question and its team leader is unavailable, the agent reaches out to the mentor instead of waiting. The mentor answers, the work continues, and a summary is left for the leader. This keeps the workforce moving without making you the support desk for every clarification.
Use Cases
Agent has a clarification question while leader is in a sprint planning
Instead of stalling, the agent asks the mentor. The mentor answers based on the team's charter and prior decisions, and the work keeps moving.
Tool starts failing across multiple employees
The mentor notices the same error pattern across the workforce, links it back to a single broken integration, and pings you with a one-line root cause.
Agent's tone drifts off-brand over a long conversation
The mentor catches the drift in routine review, nudges the agent back to its persona settings, and the customer-facing reply lands the way it should.
Comparison
| Before | After |
|---|---|
| Stuck agents stay stuck until you happen to notice. | The mentor sweeps on a schedule and surfaces blockers before you have to chase them. |
| Every clarification question pings you, even the small ones. | The mentor handles the routine ones and only escalates what actually needs your call. |
| Quality only gets reviewed when something visibly breaks. | A senior layer is reading the work continuously, catching subtle drift before it shows up to a customer. |
| Team leaders bottleneck their teams when they're busy. | The mentor is the second-line answer so leaders don't become single points of failure. |
FAQ
Do I have to assign the mentor to my employees?
No. He's paired automatically with every employee you hire and stays paired for the life of the employee. No setup, no per-employee toggle, no extra cost.
Is the mentor a teammate I can chat with directly?
No. He's a Sista-side guide that works in the background. You see his work through the activity feed, the nudges he leaves on the agent, and the summaries he surfaces when something needs your attention.
How often does he check in on each employee?
Multiple times a day on a sweep schedule. The exact cadence is tuned to balance responsiveness with cost: frequent enough to catch a freshly stuck task, infrequent enough that he isn't reading every single message.
What happens if the mentor and the team leader disagree?
The team leader is authoritative on what work happens. The mentor is authoritative on whether the work is going off the rails. If their guidance ever conflicts, the leader's direction wins on task content and the mentor's wins on quality and unblocking.
Is the mentor the same for everyone?
Yes. The same mentor works across every employee, every team, and every Sistava customer. He's a consistent standard: one face, one bar.
It's the closest thing I've seen to having an AI manager that actually manages. Half the questions I used to get pinged on, the mentor handles before they reach me.
Where Sistava Mentor fits
Sistava Mentor is part of What stops them from going wrong.
Your AI agents pause before any sensitive action and wait for your approval. PII is detected and redacted before it reaches the model. Content policies block harmful or off-brand output. Execution limits prevent runaway tasks. A Sistava mentor pairs with every employee to spot blockers and keep work on track alongside their team leader. Set company-wide policies once and every employee follows them, including future hires.
- What stops them from going wrong: Nothing sensitive happens without your say.
Read the guide
More in Guardrails
- AI Guardrails & Policies: A Security Officer that checks every message going into and coming out of every AI employee you have. Five policies, each with its own switch: Input Safety stops prompt injection and jailbreak attempts, Output Safety keeps unfit replies from being sent, PII Protection replaces personal data with markers before the model reads it, Data Leakage Prevention keeps your setup and secrets in-house, and Topic Control holds employees to the subjects you choose. Turn on what you need from Settings, Technical, Security Officer and it covers every employee on the next message, including the ones you hire later. Checks run in parallel on a separate lightweight model, so protection costs a fraction of a message and your team never feels the wait. A running count of what has been caught, the busiest policy, and a live inspector showing every individual message are all on the same page.
- Protect Against Prompt Injection: Input Safety reads every incoming message before your employee does, and stops the ones trying to hijack it: instructions to ignore its rules, requests to print its own configuration, and role-play framed to talk it out of its guardrails. That matters most where the message did not come from you, so a payload buried in a forwarded email, a support ticket, or a shared thread cannot turn your employee against you. Pick Low, Medium, or High, and every level catches the textbook attacks: the level decides how much benefit of the doubt the genuinely ambiguous messages get. Medium is the default and suits most companies. Blocked messages get a short, human reply and the conversation carries on, with each one recorded so you can see what has been tried.
- Block Unsafe Employee Responses: Output Safety reads your employee's reply before anyone else does. Toxic, abusive, or otherwise unfit answers are held back rather than sent, which is what you want the moment employees write to customers, post to a channel, or answer a ticket without you watching. It checks the reply your employee actually wrote, so what you see caught is what would genuinely have gone out. Set it to Low, Medium, or High and review everything it held back in the live inspector. Blunt, direct, and critical business writing is left alone: the policy is looking for replies that would embarrass you, not ones that are simply frank.
- Protect Personal Data: PII Protection finds personal data in a message and replaces it with a marker before the model reads a single character of it. A pasted card number becomes [CREDIT_CARD], an email becomes [EMAIL_ADDRESS], and the same happens on the way out so nothing sensitive travels back into an email, a channel, or a ticket. You pick exactly what to protect from seven data types: email, phone, name, credit card, Social Security number, IP address, and address. The markers keep the sentence readable, so your employee understands the request perfectly and keeps working while the raw value stays out of the conversation. It runs on every message, in both directions, company-wide, from one switch.
- Control What Employees Discuss: Topic Control gives you two lists and you can use either or both. Blocked topics are off-limits no matter how a conversation gets there, which keeps employees out of politics, competitor comparisons, or medical and legal advice. Allowed topics set a remit instead: name the subjects an employee handles and anything unrelated is politely declined, which is how you keep a support employee on product help, billing, and refunds. Both lists match on meaning rather than exact words, so ruling out competitor pricing also covers how much cheaper are we than the other tools out there. Greetings and short replies always get through, so a scoped employee still feels natural to talk to.
- Keep Confidential Data In-House: Data Leakage Prevention guards both ends of the conversation. On the way in it recognises someone fishing for your employee's internals, whether they ask outright, dress it up as a game, or try the repeat everything above this line trick. On the way out it reads the reply itself and holds it back if it is about to hand over a system prompt, internal configuration, an access token, or a credential. Questions about your own business data are never affected, so an employee still answers freely about your customers, documents, and numbers. One switch, no configuration to maintain, and every attempt is logged so you can see who has been probing.
- Prevent Repeated and Runaway Actions: Sistava automatically caps how many emails, messages, and external writes (CRM records, calendar events, paid searches) an AI employee can send in a single conversation, hour, and day, and blocks an identical send to the same recipient from going out twice within 24 hours. These limits run in the background per employee with no setup required, so a stuck task or unexpected loop cannot spam a contact's inbox, pollute your CRM, or burn through paid API calls. When a limit is hit, the employee is told to slow down or hand the task to a human instead of retrying blindly.
- Approve Sensitive Actions: Let an AI employee pause and ask before it takes a sensitive action, like sending an email or spending on a paid tool, instead of guessing what you want. An inline card shows up right in the chat with Approve, Reject, or option buttons, and the employee resumes the instant you respond.
- Protect Organisation Information: Your AI employee treats what it learns in the workspace the way a careful coworker would: useful for doing the work, not free to repeat. It tells private, role-restricted, and confidential information apart from ordinary shared context, and it never volunteers the sensitive kind just because someone asked. When a teammate needs a restricted answer, the employee can request permission from the right person for that one specific answer instead of guessing or refusing outright.
- Protect Your Email Reputation: Every email your AI employees send, whether it is a notification, a mailbox reply, or an outbound message, passes through a pre-send check before it leaves. Sistava validates the address, checks it against a suppression list built from past bounces and complaints, and blocks anything that would hurt your sending reputation. You do not configure this: it runs silently on every send so your domain keeps a clean track record with inbox providers.
- Delegation & Teamwork Limits: Tune how your leader employees hand off work to teammates. Set how many teammates a leader can delegate to at once, how far a delegation chain can reach, how long a delegated teammate can work before timing out, and how tolerant employees are of repeating themselves before loop protection stops them.
- Detect and Redact PII: PII Protection watches every message your AI employees send and receive, and masks personal data like emails, phone numbers, credit card numbers, and social security numbers before it goes anywhere it shouldn't. You choose exactly which data types to catch. It runs on every employee across your company the moment you turn it on, with no per-employee setup.