Meta Technical Program Manager Interview Guide

The short answer: Meta publishes its own TPM interview guide, and it defines the onsite as five 45-minute interviews: Technical Retrospective, Solution Design, Program Sense, Partnership and Behavioral. An engineer runs the Technical Retrospective, a cross-functional partner runs Partnership, and TPMs run the rest. Before the onsite, Meta describes a recruiter call, and candidate reports describe a technical screen with a Meta TPM. Meta says the whole process usually takes about 2 to 3 months. Meta treats AI tools as outside assistance unless Meta provides them, and only the tools inside the interview environment may be used. Last reviewed October 2026.

Meta technical program manager interview: quick answers

The Meta technical program manager role

What Meta technical program managers work on, the skills its job postings ask for, and what a week can look like. Meta posts the role as Technical Program Manager, often with a team name (for example Capacity Planning and Delivery, AI Infrastructure, ASIC or Ranking AI), and Technical Program Manager Intern.

Meta technical program manager levels and pay

Your level sets the bar you are interviewed against and the pay band of the offer.

Meta's TPM postings name no levels. Its TPM guide describes senior and staff expectations, and postings ask for 8+ years (Child Safety), 10+ years (the general TPM posting) or 12+ years with 8+ as a TPM and mentoring other TPMs (Capacity Planning and Delivery).

Pay in Meta's US job postings

Meta's US TPM postings each list one base salary range and mention bonus and equity on top. They name no levels, so the rows use the posting titles.

Two bands cover most postings: $168K to $234K and $199K to $272K. The Capacity Planning posting in the higher band asks for 12+ years, and the Child Safety posting asks for 8+. The intern posting, which lists monthly pay, is left out. Ranges seen October 29, 2026.

Meta technical program manager interview process

The onsite rounds, their length and what each one tests come from Meta's TPM interview guide. The recruiter call and the timeline come from Meta's hiring process page, and the technical screen comes from candidate reports.

Based on Meta's own pages and candidate reports.

Meta says each onsite interview leaves a few minutes for your questions and that the whole process usually takes about 2 to 3 months. A candidate guide labels the rounds a little differently but describes the same five-round shape and puts the process at four to eight weeks. Last reviewed October 2026.

What is different at Meta

• Meta wants TPMs who go deep. Its guide says TPMs are not crisis managers, technical consultants or communication facilitators. They dive deep into system design and architecture and act as subject matter experts.
• The interviewers are mixed. An engineer runs the Technical Retrospective and a cross-functional partner runs Partnership, so two of the five rounds are judged by the people you would work with every day.
• Solution Design blends product sense with system design. Users, requirements and KPIs come before the architecture, and Meta asks you to carry every feature you propose into the technical design.
• The bar rises with level. For Program Sense, Meta expects senior candidates to bring more complex, ambiguous programs, and says staff-level candidates often spotted an unowned opportunity and drove it. For Partnership, staff-level candidates show two-way communication and push back with leaders.
• TPMs work across the whole company. Product TPMs sit between technical execution and business strategy across Facebook, WhatsApp, Instagram, Messenger and Reality Labs, and infrastructure TPMs run capacity planning cycles for compute, storage, GPU and network capacity.
• AI shows up in the job and in the rules. Recent postings list AI-related preferred qualifications, such as using AI tools to redesign workflows. In interviews, Meta counts AI tools as outside assistance unless Meta provides them.

Meta technical program manager interview questions

Questions from our bank, then practice prompts we wrote for this guide around Meta's work.

1from typing import Dict, List, Tuple
2
3def allocate(supply: Dict[str, int], requests: List[Tuple[str, str, int, int]]) -> Tuple[Dict[str, int], Dict[str, int]]:
4    """Return (units granted per team, unmet units per team with a shortfall)."""
5    pass

1from typing import Dict, List, Tuple
2
3def allocate(supply: Dict[str, int], requests: List[Tuple[str, str, int, int]]) -> Tuple[Dict[str, int], Dict[str, int]]:
4    left = dict(supply)
5    granted, shortfall = {}, {}
6    order = sorted(range(len(requests)), key=lambda i: (requests[i][3], i))  # stable on ties
7    for i in order:
8        team, region, units, _ = requests[i]
9        if region not in left:
10            raise ValueError(f"unknown region {region}")
11        if units < 0:
12            raise ValueError("negative units")
13        give = min(units, left[region])
14        left[region] -= give
15        granted[team] = granted.get(team, 0) + give
16        if units > give:
17            shortfall[team] = shortfall.get(team, 0) + units - give
18    return granted, shortfall
19
20# Complexity: O(n log n) time for the sort, O(n + r) space for n requests and r regions.

How to prepare

A plan for the Meta loop, most important first:

1. Read Meta's TPM interview guide and plan your prep around its five rounds: Technical Retrospective, Solution Design, Program Sense, Partnership and Behavioral.
2. Practice Solution Design on a whiteboard with Meta-style products (chat, feed, video, location, reservations, shopping), product clarity first and architecture second.
3. Pick your Technical Retrospective project, reread its design docs and rehearse the 5, 2, 25, 5 minute shape, including one production problem.
4. Build Program Sense stories that cover the full lifecycle, from kickoff and buy-in to risks and communication, at the complexity your level calls for.
5. Prepare Partnership and Behavioral stories for each of Meta's five signals, including a time you pushed back on a leader.
6. Do mock interviews with no outside help, since Meta counts AI tools as outside assistance unless it provides them.

Practice path

Courses that cover the rounds above:

• System Design
• Product Sense
• ML System Design
• Product Opportunity Sizing