# How to manage a forward deployed engineer

**How to manage a forward deployed engineer** starts with a different contract than a desk IC.
          FDEs live between customer urgency and platform integrity.
          Your job is to make that dual load sustainable, visible, and fair in performance talks.
        

        

          If you still need the role definition, start with
          [what is a forward deployed engineer](https://isilta.com/blog/what-is-a-forward-deployed-engineer/),
          then come back to the operating model below.
        

        
## Clarify dual accountability in writing

        

          FDEs often hear two masters: the account wants speed, the platform team wants reuse.
          Write the decision rights. Who can approve a one off? Who owns severity when production breaks on site?
          Ambiguity here creates either abandoned customers or a fork of your product.
        

        

          Put the contract in the team working agreement and revisit it when a new logo lands.
          Career conversations should reward people who protect the platform while still delivering the account.
        

        
## Design a cadence that survives travel and embeds

        

          Weekly 1 on 1s still matter, but the format must flex.
          When someone is embedded, keep a short written pulse mid week and a deeper call when they return.
          Do not wait for them to volunteer exhaustion. Ask about sleep, cognitive load, and whether the customer is treating them like free staff augmentation.
        

        

          - Protect recovery days after multi week embeds.

          - Ban stacking critical on call with heavy travel unless you have a real backup plan.

          - Keep a shared doc of open customer risks so context survives if they go offline.

        

        
## Set performance criteria that match the work

        

          Counting pull requests alone undervalues the job.
          Count outcomes the company actually needs: time to first production value, reduction in escalations, reusable assets upstreamed, and quality of field insight that changed roadmap priorities.
        

        

          In calibration, bring quotes from customers and from core eng peers.
          Otherwise FDEs look either like heroes or like people who never ship enough core product.
        

        
## Stop the custom services trap

        

          The classic failure mode: every customer gets a special path, the FDE becomes the only person who understands it, and the platform stagnates.
          Create a weekly exception review. Any glue that lives longer than a sprint needs an owner on the product roadmap or an explicit kill date.
        

        

          Celebrate upstreaming in the same breath as customer wins.
          If promotions only reward firefighting, you will train firefighters.
        

        
## Coach the soft power without turning them into account managers

        

          FDEs need facilitation and stakeholder skills, but they are still engineers.
          Help them set boundaries with sponsors, escalate scope creep early, and write crisp decision memos.
          Do not silently convert them into pure project managers who never get deep time.
        

        
## Watch burnout early

        

          Context switching, travel, and emotional labor with frustrated operators add up.
          Use the same seriousness you would for
          [burnout signals in engineers](https://isilta.com/blog/signs-of-burnout-in-engineers/):
          shorter temper, slipping quality, cynicism about customers, or a sudden drop in written updates.
        

        

          Rotate accounts when possible. Pair juniors with seniors on embeds. Budget learning time between deployments so skills compound instead of only reacting.
        

        
## A simple operating checklist

        

          - Write the dual accountability model and share it with sales and product.

          - Define reusable versus temporary work with a review ritual.

          - Run flexible 1 on 1s with energy and risk as first class topics.

          - Measure outcomes plus upstream leverage in reviews.

          - Schedule recovery and learning after heavy embeds.

## Frequently asked questions

### Who should an FDE report to?

Usually an engineering manager with a dotted line to the account or deployment lead. One person must own career growth, compensation, and load balancing.

### How do you prevent FDEs from becoming custom services?

Set a rule for what becomes product versus temporary glue. Review exceptions weekly. Reward upstreaming, not endless one offs.

### What belongs in an FDE 1 on 1?

Customer risk, personal energy after embeds, skills growth, and which learnings should change the roadmap. Status alone wastes the meeting.

### How do you measure FDE performance?

Customer outcomes plus platform leverage: reused integrations, reduced time to value, fewer fire drills, and clear written feedback to product.

Related: [what is a forward deployed engineer](https://isilta.com/blog/what-is-a-forward-deployed-engineer/), [managing engineers more senior than you](https://isilta.com/blog/managing-engineers-more-senior-than-you/), [signs of burnout in engineers](https://isilta.com/blog/signs-of-burnout-in-engineers/).