Aurora LogoThe talent graph for frontier teams
For StartupsFor CandidatesJobsPricing
Aurora Logo
  • For StartupsHire engineers, faster
  • For CandidatesFind the role worth moving for
  • JobsBrowse open roles on the job board
  • PricingPrice your next search

Tools

  • Salary BenchmarkPrice your budget against the market

technical recruiting for startups

Where frontier teams find builders before everyone else.

Aurora maps the market, predicts which engineers your offer can win, and screens them. Our team reaches the strongest and introduces only the few exceptional fits.

20k01234567890123456789k

mapped

—

screened in depth

—

past our bar

top 0.1%our barbelow
snapshot
9,0000123456789,012345678901234567890123456789Mapped companies
470012345678901234567890123456789Open roles across our network

The builders we place come from

Where they've worked

Google DeepMindMetaBending SpoonsFigmaNotionSeries C+ · $11BApplePlaidSeries C+ · $8BNetflixTeslaTikTokHightouchSeries C+ · $2.75BLlamaIndexSeries A · $27.5M raisedFlawlessSeries B · $76.5M raisedSeries C+ · $11BSeries C+ · $8BSeries C+ · $2.75BSeries A · $27.5M raisedSeries B · $76.5M raised

Where they studied

MITStanfordUC BerkeleyCarnegie MellonHarvardPrincetonColumbiaUniversity of WaterlooUniversity of TorontoUniversity of Washington
Where searches are won

We find the strongest engineers your role can realistically hire.

We build the target market from the role backwards.

01

We calculate your real hiring ceiling

Equity is not worth the same at Seed, Series A and Series C.

We compare cash, stage-adjusted equity, scope, location and company pull with candidates' existing options. This tells us which percentile of the relevant market your complete offer can compete for.

02

We search by work, not titles

The closest technical analogue may be in a completely different industry.

We break the role into systems, scale and constraints; find teams that solved comparable problems; then identify the engineers who built the relevant parts.

03

We target the top of the reachable market

Past a certain point, candidate quality rises while close probability collapses.

We evaluate both sides: evidence that someone can do the work, and evidence that this role improves their trajectory. We search at the upper edge where both remain true.

Try it on your own role

Who can your role actually hire?

If the role as written can't be filled, you hear it in week one with the evidence.

Where the role is
$270k
$180k$620k

Cash and equity together, per year — the number an engineer compares against what they already have. Our salary benchmark puts a series a company here at .

Equity

Stage

Workplace

How well known you are

0.1%0.3%1%3%5%10%25%$146k$190k$249k$322k$422k$617k$1.08Min reachour barpast this we do not goyour packageWhat it would take to move them →Where they grade for this job →

could realistically attract

19%

of everyone past our bar
in SF Bay Area

past the bar, and would move for this past the bar, out of reach top the occasional near miss past the floor, never contacted

the list most searches build

25names off a pedigree filter

Of those, 3 would move for this package. The rest are excellent, expensive, and already fielding more approaches than they read.

the list we build

15past the bar and in reach

12 of them carry no logo a filter would catch, and the middle of them would move for $234k

An illustration of the trade-off, not a readout of real candidates. Pay expectations are the market curve behind our salary benchmark, in SF Bay Area and in local currency.

How we work

Quantitative sourcing, from your budget to the offer.

Calibratewhat your offer can winMapwho exists, and who is reachableReachwhat earns a real responseAssesswhat we can prove, and what we can'tDeliveran introduction, and what comes backwho answers re-orders itpay objections re-price itfailing the same line re-aims ityour calls re-score ourswho you pass on moves the ceilingYour offer, in candidatesstage, cash, equity, cityEvery must-have, costedpriced in candidates removedWho else is in their inboxand what they are payingA search that can closeor we tell you what to changeTeams that solved itrarely in your industryThe people who shipped itcommits, papers, launchesWho might be open nowfrom recent public activityEngineers we already knowspoken to, not scrapedThe order we work infit first, then timingThe salary is in line oneno call to find outWe open with their workthe specific thing they builtYes to the role comes firstwe assess people who want itEvery reply, and the reasona no with a reason beats silenceYour brief becomes a rubricagreed with you before we startSame questions, every timeso answers can be comparedEach line answered, or notwith the evidence attachedWhat we could not verifyreported, never rounded upOur screen, scored by yoursaccuracy by requirement and sourceA written case, not a CVfrom the person who interviewedWhat could still stop itthe risks, before you meetTheir number, up frontnot after four interviewsYour call, per requirementwhich line failed, not just no
What you get

We intro everyone who clears the bar.

We don't sell you a shortlist. We keep sending strong engineers until you are satisfied, you can hire as many as you want.

One engineer at a time

Your bar7 requirements
45%Under your bar

Introduced · one at a time · with no cap

Founding engineer, applied AI

Series A · 14 people · San Francisco
87%

Strong fit

100% of the bar readable

1 never scored
Wants founding scope back
Won't manage anyone
$215–240k · free in 6 weeks

CV + interview

  • Shipped AI to customersmust

    40k paying users

  • Python + a typed languagemust

    Python + Go

  • 5+ years, startup

    7 yrs, 2 startups

  • End-to-end ownership

    Carried the pager

  • Retrieval and evals

    Built the eval set

  • Credible with customers

    Ran the design-partner calls

  • On-site 4 daysnot scored

    Wants the room

Our bar

You see the profiles the whole graph agreed on.

×NAB
  • Mapped20k
  • Read by a person—
  • Past the bar—
  • —
  • —
  • —
  • —
  • —
  • our bar
  • —
  • —
  • —

Reading…

What a CV can't tell you

We know what makes startup hires work

Not everyone is ready for a startup move. We establish that before making an introduction.

Wants this move, not just a job

Candidates already in a process learn to say yes to everything. Ours weren’t looking until we gave them a reason.

Won't leave in a year

We trace the reason behind every move they’ve made—and what would make them move again.

Owns what breaks

A strong CV can hide someone who executed well but never owned the decision.

Runs at your pace

Everyone agrees to intensity in an interview. We look at how they spent the last two years.

Will be in the room

“Open to relocation” often dies at offer stage. We verify it against real-life constraints before you invest the time.

Takes the upside

Risk aversion kills offers. We qualify the trade-off between cash, equity and certainty from the start.

“The engineers you want aren't looking. Everything we do is built around that one fact.”
The Aurora team

Our network

A few of the engineers we've met.

Jua.ai

Senior AI Researcher

Jua.ai

Research Standout

$27M raised; trained >1B-param weather models for energy trading.

NNAISENSE

NNAISENSE · Co-founder / Director of AGI

Utrecht University

Utrecht University · PhD Artificial Intelligence

AGI Author + WEF Speaker

Meta

Staff ML Engineer

Meta

Machine Learning

Develops ML models for Instagram post analysis and profile credibility.

Johnson & Johnson / Verb Surgical

Johnson & Johnson / Verb Surgical · Staff Deep Learning Engineer

UC Berkeley

UC Berkeley · M.S. Operations Research

23 Publications, 8 Patents

Apple via Ryzen Solutions

AI/ML Consultant

Apple via Ryzen Solutions

Computer Vision

Led production 3D vision and pose estimation systems adopted in Apple workflows.

Carnegie Mellon Robotics Institute

Carnegie Mellon Robotics Institute · Research Assistant

Carnegie Mellon University

Carnegie Mellon University · Graduate Robotics Research

Apple Daisy Deployment

Retell AI

Infrastructure Engineer

Retell AI

AI

YC-backed, $4.6M Seed; building infra for AI voice agents.

Oracle

Oracle · Software Engineer II

Georgia Tech

Georgia Tech · M.S. Computer Science, AI

VLDB + Learning at Scale

Seed-stage AI SRE startup

AI Research Engineer

Seed-stage AI SRE startup

Research

Built AI incident investigation agents, memory, skills, alert grouping, and GitHub workflows.

Microsoft

Microsoft · Software Engineer Intern

Georgia Tech

Georgia Tech · M.S. Computer Science

Top 0.05% IIT JEE

Meta

Senior Software Engineer

Meta

Software

Owns Horizon OS app management across device, mobile app, and backend.

Microsoft

Microsoft · Software Engineer II

Appalachian State University

Appalachian State University · B.S. Computer Science

Plaid

Software Engineering Intern

Plaid

Infrastructure

Building Go/Kubernetes networking infra serving 8,000+ financial institutions.

Figma

Figma · Software Engineering Intern

UCLA

UCLA · B.S./M.S. Computer Science

TaxGPT

Research Scientist

TaxGPT

Research

YC-backed, $4.6M Seed; 4th eng hire building AI tax agents.

W

Womp Labs · Machine Learning Intern

Minerva University

Minerva University · B.Sc. Computer Science: Data Science

What we recruit for

We focus on a few categories.

Founding Engineers

Product generalistSystems builderEx-founder

What they shipped before anyone wrote a spec.

Being early at a startup.

AI Engineers

Pre-trainingPost-trainingInferenceApplied

The bottleneck they moved: training stability, tokens/sec, cost per call.

Experience with LLMs.

Eval Engineers

BenchmarksHuman evalAutomated judgesRed teaming

An eval that changed a training run or stopped a launch.

A dashboard nobody acts on.

Infra Engineers

Dev velocityReliabilityDistributed systemsCloud cost

A number they moved: p99, cloud spend, time to deploy.

Kubernetes on a CV.

Backend Engineers

APIs & servicesData & storagePaymentsAsync & queues

A migration or API change they shipped under live traffic.

Scale they inherited.

Full-Stack Engineers

ProductFrontend-heavyBackend-heavyGrowth

One thing they took from idea to adoption themselves.

A list of frameworks.

Computer Vision Engineers

Detection & tracking3D & SLAMImagingEdge deployment

Noisy labels, drifting scenes, hardware they couldn't change.

A leaderboard score.

Research Engineers

Reproduce & extendScalingPerformanceDomain depth

What the experiments that failed taught them.

Citation count.

Pricing

A percentage of the hire. Nothing hidden.

Retainer is where we do our best work, and where most teams start.

Most teams start here

Retainer

12–25%of salary

20% upfront, credited against the final fee.

  • Prioritized, dedicated search
  • Priced by search complexity
  • Best outcomes and speed

Contingency

15–30%of salary

No upfront. You pay only when you hire.

  • Only pay when you hire
  • Higher rate for the added risk
  • Good for exploratory searches

Have a role that's hard to fill?

Aurora Logo

The infrastructure that helps builders find frontier teams

Terms of Service•Privacy Policy

© 2026 Aurora Jobs ApS. All rights reserved.

Aurora Jobs ApS — Registered office: Ved Lunden 4, C/O Blue Office, 8230 Åbyhøj, Denmark • CVR 46223780
Operational office: C/o CSE (Copenhagen School of Entrepreneurship), Porcelænshaven 26, 2. sal, 2000 Frederiksberg, Denmark