Bullets written from what you actually did

Describe what you actually did in your own words, even if it is rough or unfinished. AI turns it into clear, compelling resume bullet points while keeping your achievements accurate and never inventing experience.

The blank bullet is where resumes die. People know what they did and cannot make it sound like anything, so they write "responsible for managing the team" and move on. What we generate is not better prose for its own sake. It is the same information in the structure that survives a six-second scan: result first, mechanism second, scale attached.

The constraint that matters is what it will not do. It will not add a percentage you did not mention, invent a team size, or upgrade your title. If a bullet needs a number to be credible and you have not given one, it asks you for it. An impressive resume you cannot defend in an interview is worse than an honest one.

In short

  • Keeps your numbers; never invents new ones
  • Three to five options per generation, not one
  • Rewrites for tone, length and seniority
  • Version snapshot taken before any bulk edit
Priya Raghunathan
Senior Backend Engineer, distributed systems and payments
priya.r@hey.com · +1 (415) 555-0142 · Oakland, CA · linkedin.com/in/priyaraghunathan · github.com/priyar · priyar.dev
Summary

Backend engineer with eight years on high-throughput payment infrastructure. Took a ledger service from 400 to 9,000 writes per second without a consistency regression, and led the migration that moved 42 million accounts off a single Postgres primary. Comfortable owning a system end to end: schema design, rollout, the pager.

Experience
Senior Backend Engineer, Northwind Payments · Remote (Oakland, CA)Mar 2022 – Present
Rebuilt the double-entry ledger on an append-only event store, lifting sustained throughput from 400 to 9,200 writes per second while keeping balance reconciliation exact.
Led the shard migration of 42M accounts off a single Postgres primary across six weeks with zero customer-visible downtime and no rollback.
Cut p99 settlement latency 71% (2.4s to 690ms) by replacing synchronous fan-out with a batched outbox consumer.
Mentored four engineers; two were promoted to senior within eighteen months.
Backend Engineer, Cartogram · San Francisco, CAJun 2019 – Feb 2022
Designed the tenancy model behind the enterprise tier, which closed $3.1M in ARR that had been blocked on data isolation.
Reduced infrastructure spend 34% ($780K annualised) by moving cold tiles to object storage behind a signed-URL cache.
Wrote the incident review process still used company-wide; mean time to resolution fell from 4.1 hours to 52 minutes.
Software Engineer, Helix Robotics · Berkeley, CAAug 2017 – May 2019
Built the telemetry pipeline ingesting 1.2TB per day from 3,400 field units, replacing a batch job that ran twelve hours behind.
Shipped the fleet-diagnostics API consumed by every downstream team and by two customer integrations.
Skills
Languages: Go, TypeScript, Python, SQL
Infrastructure: Kubernetes, Kafka, Postgres, Terraform, AWS
Practices: Distributed systems, Event sourcing, Observability, Incident command
Education
B.S. Electrical Engineering & Computer Science, University of California, Berkeley · Berkeley, CAAug 2013 – May 2017
Grade: 3.8/4.0

How it works

  1. 01

    You describe, in any state

    "I fixed the checkout thing that kept timing out, took it from like 2 seconds to under a second, took me about a month" is a perfectly good input.

  2. 02

    It returns three to five options

    Different emphases (one leading on the metric, one on the scope, one on the mechanism), so you are choosing rather than accepting.

  3. 03

    You edit, then it fits

    Rewrite any line shorter, longer, more senior, or less technical. Length control matters more than people expect: a bullet that wraps to a third line costs you a bullet elsewhere.

What it writes besides bullets

A professional summary from the rest of the document, so it actually reflects the resume underneath it rather than the aspirations of whoever wrote the template. Skills suggestions for a target title, drawn from what your experience already implies. Cover letters that open on something specific from the posting. Resignation letters, for the part of the process nobody wants to draft.

It also translates a whole document into another language while leaving the layout, dates and structure untouched, which is the difference between applying abroad and rebuilding your resume from scratch.

Why generation is metered and scoring is not

Every generation is a model call with a real cost, so it draws on a monthly credit allowance: one credit for a set of bullets, three for a cover letter, five to tailor a whole document. Scoring, editing, exporting and switching templates cost nothing, on every plan.

The free plan includes ten credits, which is enough to write a summary, generate bullets for two roles and see whether the output is worth paying for. That is the intended use.

Questions

AI writing questions

Will an employer know it was written with AI?

They will know if it reads like nobody was there: no specifics, no numbers, the same three adjectives everyone uses. Written from your real work with your real figures, it reads like a resume with a good editor, which is what it is. Edit the output; that is what the field is for.

Which model is behind it?

Llama 3.3, running on Cloudflare's network rather than a third-party API, so your text is not handed to another vendor to get a bullet back. The reply is constrained by the document's own schema during decoding and validated against it afterwards, so it arrives as structure rather than prose we have to parse.

Can it make things up if I ask it to?

It will not add metrics or claims you have not provided. If you type a false number into the input yourself, it will use it. The honesty of the input is yours to own.