Building an Engineering Team by Ammon Bartram and Harj Taggar

Foundations of Engineering Hiring

• Hiring engineers is identified as the primary bottleneck for YC alumni and startups, characterized by high time costs, repetitive tasks, and emotional rejection.
• The hiring process is modeled as a funnel consisting of three distinct stages: sourcing (top), screening (middle), and closing (bottom).
• Founders are advised to exhaust personal networks first, as early-stage success heavily depends on the ability to work effectively with co-founders and early employees.
• A "sourcing party" is recommended where the team brainstorms potential candidates together using shared spreadsheets to map their networks against role requirements.
• Referral bonuses and dedicated team events are suggested to incentivize current employees to leverage their personal networks for future hires.

Sourcing Channels Ranked by Priority

• Personal Networks: Ranked as the most effective source for early-stage hires because they reduce the risk of cultural and working-style mismatch.
• Hiring Marketplaces (e.g., Triple Byte, Hired, Vetri):
  • Function similarly to dating sites where candidates have more power due to high demand for engineering talent.
  • Offer speed of hire as candidates are actively looking, though competition is high.
  • Typically charge 15–20% of the first year's salary; Triple Byte reports a 2x higher offer rate compared to other sources.
• LinkedIn and GitHub:
  • Traditional cold outreach via these platforms suffers from declining response rates due to high recruiter volume.
  • Founders should avoid high-volume spamming and instead send personalized, short messages that demonstrate deep profile research.
  • Sending direct emails (via tools like Connectifier) yields significantly higher response rates than platform messaging.
• Job Boards (Stack Overflow, AngelList, HackerNews):
  • Suffer from a quantity-over-quality issue, often flooding companies with underqualified applicants.
  • Success requires unique, personality-driven job descriptions that avoid corporate boilerplate and highlight specific technical challenges.
  • HackerNews Jobs is noted as having high quality but is restricted to YC companies.
• In-Person Meetups:
  • Generally considered ineffective for immediate hiring due to low attendance of active job seekers.
  • Technical meetups (e.g., local coding groups) or hosting office meetups are recommended over general industry conferences.

Recruiter Engagement Strategy

• Founders should wait until they have hired their first engineer and understand the specific pain points of the process before delegating to a recruiter.
• Hiring a recruiter is recommended when sourcing activities consume over 50% of a founder's time.
• A hiring cadence of roughly one engineer per month is suggested as a baseline before bringing on full-time help to ensure utilization.
• Contract Recruiters: Recommended as the first step to handle sourcing and calendar filling while founders handle pitching.
• In-House Recruiters: Considered once the founder needs to delegate the pitching and initial screening process.
• Agencies: Charged 25–30% of first-year salary; noted for being high-touch but often sending the same candidates to multiple companies.

Reducing Noise in Technical Screening

• Data on Interview Consistency: Inter-rater reliability among interviewers is extremely low (approx. 0.1), statistically similar to Netflix user agreement on movies, highlighting high "noise" in the process.
• Define Critical Skills: Companies must explicitly decide which skills matter (e.g., speed vs. rigor, product focus vs. deep tech) to prevent interviewers from biasing against candidates based on irrelevant criteria.
• Structured Interviews:
  • Unstructured interviews are less predictive than structured ones where every candidate is asked the same set of questions.
  • Interviewers should be given defined criteria to evaluate specific skills rather than making global "gut feeling" decisions.
  • Decision-making should be centralized to a single person or group reviewing all structured notes.
• Question Design:
  • Avoid "leaps of insight" questions (e.g., Fibonacci stairs) that can be guessed; prefer multi-step problems like implementing Connect Four.
  • Budget 3x the time it takes an expert to solve a problem for the interview slot, as interviewers often underestimate question difficulty.
  • Ask 4+ questions per session to average out the noise of individual answers.
• Credential Blindness: Interviewers should not see resumes or credentials (e.g., school, past employers) during technical evaluations to prevent bias; credentials can be considered later in the decision meeting.
• False Negatives vs. False Positives:
  • Both error types are costly; startups must balance the risk of passing a bad hire against the risk of missing a good one.
  • Data shows no engineer passes all interviews, and many who fail one process succeed elsewhere, suggesting hiring bars may be too high.
• Calibration: Evaluate candidates based on their maximum skill potential rather than their weakest link, unless the weak area is critical to the job.
• Candidate Experience:
  • Stress reduces performance; allowing candidates to use their own laptops, languages, and tools improves results.
  • Interviewers should be trained to be supportive; a positive experience increases offer acceptance even for rejected candidates.
  • Avoid "hazing" interviews where candidates are treated as outsiders.

Early-Stage Specific Tactics

• Small Teams: If unable to standardize the process fully, still define core skills and use structured criteria to reduce bias.
• Sourcing Bottlenecks: If applicant volume is low, prioritize minimizing false negatives by screening less aggressively before onsite interviews.
• First Hires: Prioritize "productivity" and "ownership" over code quality for the first few hires to ensure rapid execution.
• Non-Technical Founders: If lacking technical expertise, assess candidates by asking them to explain their solutions in terms the founder can understand.

Closing and Offer Management

• Speed as a Moat: Startups have a distinct advantage in speed; delaying offers allows candidates to accept competing offers.
• Interviewer Behavior: Complaints from candidates often stem from unprepared interviewers or those trying to intimidate the candidate rather than evaluate them.
• Culture Pitching:
  • Move beyond generic adjectives like "open, transparent, collaborative."
  • Differentiate by discussing trade-offs (e.g., "openness means difficult honesty").
  • Be prepared to answer specific questions about diversity plans.
• Team and Investor Involvement: Engage the team and investors with strong closing skills to reinforce the candidate's decision to join.
• Transparent Offers: Provide full compensation details immediately; avoid "bait and switch" tactics or hypothetical negotiation questions.
• Equity Communication: Ensure founders understand stock options and tax implications to effectively communicate value to candidates unfamiliar with equity.

Competing with Tech Giants

• Learning and Responsibility: Pitch the speed of learning via real decision-making authority, which is restricted in large corporations.
• Career Progression: Emphasize the meritocratic nature of startups where rapid promotion is possible without paying "dues" over decades.
• Opportunity Cost: Frame the startup role as a unique, non-fungible opportunity that won't be available again, contrasting with the generic nature of big tech roles.
• Mentorship: For junior hires, emphasize the opportunity to learn from experienced engineers on the team who have worked at major tech firms.

Q&A Insights

• Background Checks: Generally not a focus for startups; evaluating past work portfolios is often more effective.
• Grades and Metadata: Triple Byte focuses on "background-blind" hiring, prioritizing direct skill assessment over correlated metrics like GPA or personality tests.
• University Pipelines: College hiring is less viable for startups due to timing mismatches and the need for immediate hires versus large signing bonuses offered by big tech.
• Mock Interview Exercise: A training method where one person role-plays a candidate intentionally making mistakes forces the interviewer to be critical, revealing their own blind spots and reducing overconfidence in their evaluation skills.
• Hiring Success Rate: Data indicates 5% of hires are fired within months; an additional 30% remain employed but are not top performers, leaving roughly 10% as true top performers.