$150K aws credits · series-a canonical

$150K in AWS credits for Series-A startups — the modal target, the approval math, the trust signals.

$150K is the credit number Series-A teams should aim for. It composes from three stacked pools — Activate Portfolio ($100K) + Build for Startups ($25K) + Bedrock POC ($25K) — and Series-A is the only funding stage where the full stack lands at the documented ceiling with high probability. This page covers why Series-A clears the AWS reviewer math, what specifically tips an application to full approval, the 18–21 day timeline, and the trust signals reviewers weight at Series-A specifically.

target stack
$150K
series-a approval rate
~80%
time-to-balance
18–21 days
cost to you
$0
TL;DR
  • Series-A is the only funding stage where the $150K stack — Activate Portfolio ($100K) + Build for Startups ($25K) + Bedrock POC ($25K) — clears AWS reviewer math at the full ceiling with ~80% probability. Seed-stage applicants attempting the same stack approve at ~35%; pre-seed at under 15%.
  • The reviewer math at Series-A: 18–24 months of projected AWS consumption sits between $120K and $250K, which exactly absorbs a $150K credit pool. Reviewers calibrate against this. At seed, projected consumption is $30K–$60K — a $150K pool overshoots and gets downgraded.
  • Trust signals reviewers weight at Series-A: institutional lead, board composition, audit-stage corporate posture, and a stack that contains a discrete AI workload. Each adds independent weight to the approval decision. Most full-ceiling Series-A approvals carry three of the four signals.
the stage fit

IWhy $150K is the modal target for Series-A — and a mismatch at other stages

Credit ceilings are not eligibility lines. They are reviewer calibration targets. The $150K stack lands at full ceiling for Series-A because the projected AWS consumption over 18–24 months sits in a window where $150K is the natural fund-to-graduation amount. At adjacent stages, the same stack lands partial.

A Series-A team with 8–18 engineers, a live product, and 12–24 months of post-round runway will burn AWS at $5K–$15K/month over the credit window. Compound that across 24 months and the total spend lands between $120K and $360K. A $150K credit pool funds 14–28 months of that spend at the median — long enough to ride past the credit window and graduate into a paying-customer relationship. AWS reviewers know this math because they built the ceiling around it.

Seed-stage teams burn AWS at $500–$2,500/month. Across 24 months, that totals $12K–$60K. A $150K credit pool is 3–10x larger than the team's actual consumption — meaning most of the credits would expire unused. Reviewers see this and downgrade the Portfolio layer to $50K or $25K, killing the stack math. Filing the full $150K stack at seed without strong AI traction is a structural mismatch.

Series-B teams are on the opposite side. Their AWS consumption is already $20K–$80K/month. A $150K credit pool funds 2–7 months. Reviewers approve, but the value proposition is weaker — Series-B teams are better routed to MAP or EDP, where the credit pool scales with consumption rather than being capped at $150K. Filing the standard $150K stack at Series-B is the right move only when the team has not yet outgrown it.

The $150K stack is exactly fit-for-purpose at Series-A. Outside that band, either the credit pool overshoots projected consumption (seed) or undershoots it (Series-B). The reviewer calibration is the constraint that makes Series-A the modal target.

reviewer calibration

IIThe AWS reviewer math at Series-A, decomposed

When an AWS reviewer evaluates a partner-filed Series-A credit application, they run an implicit calculation: projected consumption versus credit pool requested versus account-graduation probability. The math determines whether the application lands full, partial, or downgraded. Here is what they actually weigh.

Variable 1 — projected monthly AWS spend

What reviewers look for at Series-A: a credible projection in the $5K–$15K/month range over the next 24 months. Projections below $4K/month signal seed-stage scale; projections above $20K/month signal Series-B scale. Either triggers a re-categorization.

How partners document this: the ACE record's "Projected Monthly Spend" field. The partner derives the number from the team's current cloud spend (Heroku, GCP, Azure, current AWS) plus expected workload growth from the funded plan. The number is defensible — reviewers occasionally ask for the basis.

Failure mode: claiming $25K/month with no supporting context. Reviewers flag and either downgrade to a believable number or reject for "implausible projection." The fix: project conservatively. $8K/month is more credible than $15K/month and produces the same $150K approval.

Variable 2 — funding signal and round age

What reviewers look for: a Series-A round closed within the last 18 months from an institutional lead. Top-tier VCs (a16z, Sequoia, Bessemer, Greylock, Founders Fund, Index, Accel, Lightspeed, GV, Khosla, NEA, Insight, General Catalyst, Tiger, Coatue) trigger immediate trust. Tier-two funds (regional, sector-specialist) trigger moderate trust. Solo capitals and angel syndicates trigger minimal trust without supplementary signals.

Round age matters: a round closed 0–12 months ago is a strong signal. 12–18 months still works but reviewers want to see continued operational health. Past 18 months and reviewers begin asking whether the company has raised again or is approaching runway end.

Failure mode: filing the stack 22 months after the Series-A close without a follow-on round. Reviewers downgrade Portfolio to $50K because they don't want to fund a company approaching zero runway. The fix: file within 12 months of the round, or include a clear path-to-Series-B narrative in the application.

Variable 3 — workload discreteness for layer-2 and layer-3

What reviewers look for: three genuinely distinct use cases mapping to the three credit pools. Portfolio for general infrastructure; Build for Startups for a specific new project; Bedrock POC for a defined AI workload with eval methodology. Three independent paragraphs, each pointing to a different AWS service surface.

The discreteness test: reviewers cross-reference the AWS services listed in each ACE record. Portfolio listing "ECS + RDS + S3 + CloudFront" alongside Build for Startups listing "ECS + RDS + S3 + CloudFront" is overlap. Reviewers approve Portfolio and zero out Build. The fix: Build for Startups should describe a discrete project using a different service surface — MediaConvert, IoT Core, Glue, Kinesis, OpenSearch.

Failure mode: framing the same product twice. "We are building a B2B SaaS" in Portfolio and "we are launching a B2B portal" in Build for Startups gets flagged. The portal has to be a genuinely different product line — a new vertical, a new customer segment, a new technical surface.

Variable 4 — AI workload genuineness for Bedrock POC

What reviewers look for: a Bedrock POC plan with four explicit components — the use case (concrete, not exploratory), the chosen model with reasoning, the eval methodology with sample size, the inference budget. Plans missing any one component get downgraded to the $10K floor; plans with all four typically land at $25K.

The model-choice signal: "Claude Sonnet 4.5 because it balances cost and quality for our scope" reads as serious. "We will use whatever model is cheapest" reads as unfunded. Reviewers parse this language directly. The chosen model also has to fit the use case — picking Nova for high-stakes legal review reads as cost-cutting in a way reviewers downgrade.

The eval methodology signal: "we will measure accuracy on N=500 held-out examples per week, comparing against a baseline classifier" lands. "We will iteratively improve" gets downgraded. Eval methodology is the single most predictive signal of Bedrock POC full-ceiling approval at Series-A.

the application pattern

IIIThe three-application stacking pattern, executed by a Series-A partner

AWS reviewers approve three separate ACE records — not a single combined application. Understanding the filing sequence and the timing between submissions is what makes the stack land in 18–21 days rather than 35–60.

Filing day — same business day, three records. The partner submits Portfolio, Build for Startups, and Bedrock POC as three independent ACE opportunity records, all on the same business day. Same-day filing is structurally important: it positions the three applications as a planned stack rather than sequential scope expansion. AWS reviewers reading three records filed on Day 5 treat them differently than three records filed across three weeks.

Use case isolation across records. Each record uses a different "Customer Use Case" paragraph. The Portfolio record describes general infrastructure for the product the company built with its Series-A capital. The Build for Startups record describes a specific scoped project — a new feature, a new vertical, a new compliance workload. The Bedrock POC record describes the AI evaluation plan exclusively, with no general infrastructure language.

Different reviewer queues, different timelines. Portfolio routes to the standard ACE reviewer queue (typically 8–12 days). Build for Startups routes to the same queue (10–14 days). Bedrock POC routes to a separate Bedrock-team-adjacent queue (12–18 days). Filing all three on the same day means the slowest queue (Bedrock POC) becomes the binding constraint. The 18–21 day total reflects this.

Approvals arrive separately, not as a batch. Portfolio typically lands first (Day 10–12), Build for Startups second (Day 12–14), Bedrock POC last (Day 14–18). Each arrives as its own email with its own credit-balance line item in the AWS billing console. Founders sometimes miss the second or third approval because they stop checking after the first. CloudRoute partners track all three to completion.

  • Pattern 1 — same-day filing — All three ACE records submitted within the same business day. Establishes the planned-stack signal rather than the sequential-scope-expansion signal that reviewers downgrade.
  • Pattern 2 — independent use case paragraphs — Each record describes a different workload using a different AWS service surface. No copy-paste between records. Reviewers explicitly check for overlap.
  • Pattern 3 — explicit eval plan in Bedrock POC — Use case + model + eval methodology + budget + window. All four components present. The single most reliable predictor of full-ceiling Bedrock POC approval.
  • Pattern 4 — projection consistency across records — The $5K–$15K/month Portfolio projection aligns with the $1K–$2K/month Build projection and the $1K–$2K/month Bedrock projection. Total monthly $7K–$19K. Numbers that don't add up cleanly trigger reviewer questions.
tipping the decision

IVWhat tips a Series-A $150K application to full approval vs partial downgrade

Series-A applications rarely get fully rejected. They get downgraded. Understanding the downgrade patterns — and what tips the decision the other way — is what separates a $150K outcome from a $75K outcome.

The most common downgrade pattern at Series-A: Portfolio approves at full $100K, Build for Startups gets downgraded to $10K because reviewers flagged the use case as overlap, Bedrock POC lands at $25K. Total: $135K instead of $150K. The $15K gap is meaningful — it represents 1–2 months of additional runway at typical Series-A burn.

The second most common downgrade: Portfolio approves at $100K, Build for Startups lands at full $25K, Bedrock POC gets downgraded to $10K because reviewers flagged the POC plan as underspecified. Total: $135K. Same gap, different cause. The fix is structural — adding eval methodology and explicit budget to the POC plan.

Full-ceiling approval requires three independent decisions to land correctly. Each has its own risk profile, its own reviewer judgment, and its own failure modes. Treating the stack as a single application underweights the work required to clear each independently.

Signals that tip Portfolio to full $100K

Tier-1 institutional lead in the round (a16z, Sequoia, Bessemer, Greylock, Founders Fund, Index, Accel, Lightspeed, GV, Khosla, NEA, Insight, General Catalyst). Reviewer trust calibration jumps when a top-15 fund's name appears in the application.

Board composition with at least one independent technical advisor or operating partner. Reviewers occasionally check LinkedIn footprints of named board members; serious advisors strengthen the application.

Audit-stage corporate posture — SOC 2 Type I or Type II in progress, terms-of-service published, privacy policy current. Reviewers cross-check these as signals of operational maturity.

A clearly itemized AWS service surface. "We will use ECS, Aurora, S3, CloudFront, CloudWatch, X-Ray, and Cognito" reads as a team that has thought through the infrastructure. "We will use AWS services as needed" reads as a team that has not.

Signals that tip Build for Startups to full $25K

A discrete new project that is genuinely orthogonal to the Portfolio use case. Adding a video pipeline using MediaConvert, building a B2B compliance portal using Cognito + API Gateway + Step Functions, launching a data pipeline using Glue + Athena + S3, or adding a real-time analytics layer using Kinesis + OpenSearch.

A 6–12 month implementation window stated explicitly. Reviewers want to see a temporal scope rather than an open-ended commitment.

A $1K–$2K/month projected spend for the discrete project, distinct from the Portfolio projection. The math has to add up coherently.

A named project owner in the application. "Engineering team will execute" is weaker than "VP Engineering Sarah Chen will lead the implementation." Named owners signal commitment.

Signals that tip Bedrock POC to full $25K (and the path to $50K)

A concrete use case described in one paragraph — "we are evaluating Bedrock to power a customer-support summarization workflow processing 80K tickets/month." Avoid exploratory language.

A specific chosen model with one-sentence reasoning — "Claude Sonnet 4.5 for the cost/quality balance at our scope." Demonstrates that the team has compared options.

An eval methodology with sample size and baseline — "N=500 held-out examples per week, measured against the current rule-based classifier with manual review of disagreements."

An explicit inference budget and POC window — "$1,200/month projected Bedrock spend, 60-day POC window with go/no-go at Day 45."

For the $50K tier (uncommon at Series-A but achievable): a Letter of Intent from a paying customer who will fund the AI feature, projected Bedrock spend of $4K+/month, demonstrated traction signal in the eval plan.

reviewer-trust signals

VSeries-A specific reviewer-trust signals and how partners surface them

AWS partners filing Series-A applications have learned which application elements reviewers notice. The signals stack — one or two adds modest weight, three or four flips a downgrade to full approval. CloudRoute partners surface these explicitly during the application discovery call.

series-a reviewer-trust signals and their weight in the approval decision · 2026
SignalHow it surfacesApproval weightFailure if absent
Tier-1 institutional leadNamed in the funding-history paragraph; cross-referenced via Crunchbase by reviewerHighPortfolio may downgrade to $75K
Round closed within 12 monthsDate stated in the round-history fieldHighReviewers ask about follow-on plans
Engineering team 8–18 strongTeam-size field in the company info sectionModerateHeadcount under 8 flags as too early-stage
Defensible AWS spend projectionSpecific $K/month figure with supporting Heroku/GCP/current-AWS contextHighVague projections downgrade Portfolio
SOC 2 in progressOne sentence in the corporate posture paragraphModerateLighter compliance signal; not blocking
Discrete AI workload for Bedrock POCStandalone use-case paragraph with model, eval, budgetHighBedrock POC drops from $25K to $10K floor
Named technical advisor on boardListed in board composition paragraphModerateNot blocking; adds weight to full-ceiling decision
Itemized AWS service list per recordBullet list in each ACE record's service sectionHighReads as not-yet-architected; downgrades signal
Series-A applications carrying 5+ signals approve at full ceiling ~85% of the time. Applications with 2–3 signals approve at full ceiling ~55%; partial $125K outcomes are common. Applications with 0–1 signals routinely approve at the floor ($65K–$85K total).
stage comparison

VISeries-A versus seed-stage attempting the same $150K stack

Seed-stage founders sometimes hear that the $150K stack is achievable and file the same applications. The approval rate is materially lower. Understanding why illustrates the reviewer calibration math more clearly than any single Series-A application.

A seed-stage team with 3–6 engineers and a $1M–$3M round burns AWS at $400–$1,800/month. Over 24 months that totals $9K–$45K. A $150K credit pool is 3–17x the projected consumption. Reviewers see the gap and downgrade Portfolio from $100K to $25K or $50K, citing "projected consumption insufficient for ceiling." The downgrade is structural, not discretionary.

Even when seed-stage applicants file with strong AI workload signals — a defensible Bedrock POC plan, clean eval methodology, tier-1 institutional lead — the Portfolio downgrade still happens. Reviewers may approve the Bedrock POC layer at full $25K (the AI signal is independent of consumption math), but the Portfolio downgrade caps the total stack around $60K–$100K rather than $150K.

The Build for Startups layer compounds the seed-stage problem. Build for Startups assumes the company has a distinct second workload at production scope. Seed-stage teams typically have one product. Reviewers flag the Build use case as either fictional (reads as padding) or as overlap with Portfolio. The default action is to zero out Build for Startups.

The honest path for seed-stage: target $50K–$80K via Portfolio at the seed-tier ceiling plus a defensible Bedrock POC. Save the full $150K stack for the Series-A close, when the consumption math will land cleanly.

series-a versus seed attempting the $150k stack · 2026 approval data
VariableSeries-A applicantSeed-stage applicant
Projected monthly AWS spend (typical)$5K–$15K$400–$1,800
24-month consumption math$120K–$360K$9K–$45K
Portfolio ceiling reviewers approve$100K (full)$25K–$50K (downgraded)
Build for Startups approval rate~75% at full $25K~25% at full $25K
Bedrock POC approval rate~80% at $25K~60% at $10K–$25K
Median stack outcome$135K–$150K$45K–$80K
Time-to-balance18–21 days14–28 days (variable)
Stack math fitCalibratedOvershoots by 3–10x
CloudRoute pipeline data, 2026 H1. Seed-stage applicants pursuing the $150K stack should expect $45K–$80K total credits. Series-A applicants pursuing the same stack should expect $135K–$150K when the application carries the trust signals described in Section V.
the 21-day timeline

VIIThe 18–21 day timeline for a Series-A $150K stack

Day 0 — CloudRoute inquiry submitted. Indicate the target is the full Series-A $150K stack. 3 minutes of founder time.

Day 1 — Routed within 24 hours to an Advanced or Premier-tier partner with Series-A track record. Calendly link in inbox.

Day 2–3 — 45-minute discovery call. Partner walks through three application planning blocks: (1) Portfolio use case + service list + projection (~15 min); (2) Build for Startups discrete project identification (~10 min); (3) Bedrock POC use case + model + eval + budget (~20 min). Founder leaves the call with a clear application spec.

Day 3–5 — Founder provides: round details + lead investor name, AWS account ID, current cloud spend documentation (Heroku/GCP/Azure bill or current AWS Cost Explorer screenshot), deck, three independent use-case paragraphs, eval methodology paragraph for Bedrock. Total founder time: ~30 minutes.

Day 5 — Partner files all three ACE records on the same business day. Portfolio first, Build for Startups second (same hour), Bedrock POC third (same hour). Same-day filing positions the stack as planned.

Day 8–12 — AWS standard reviewer queue assigns. Portfolio typically lands first — approval email arrives Day 10–12. Credit balance shows in the billing console within 24 hours of the approval email.

Day 12–14 — Build for Startups approval typically lands second. Second credit balance line item appears in billing console. Distinct from Portfolio in expiration date (12 months vs 24).

Day 14–18 — Bedrock POC approval typically lands last. Bedrock-team-adjacent reviewer queue runs 4–7 days slower than standard. Credit balance shows as tagged for Bedrock services with a 60-day POC checkpoint.

Day 16–21 — All three balances confirmed in the AWS billing console. Total $150K applied. Partner closes the engagement and provides written confirmation of the credit composition for the founder's records.

Day 21+ onward — Credits auto-apply to monthly AWS invoices in expiration order. At typical Series-A burn ($6K–$8K/month effective consumption), the stack lasts 18–24 months. Bedrock POC pool burns first (12-month validity); Build for Startups second (12-month validity); Portfolio last (24-month validity).

why same-day filing matters

Filing the three ACE records on three different days reads to reviewers as scope expansion of an already-approved engagement. AWS reviewers downgrade the second and third applications because the assumption is that the first application captured the workload. Same-day filing positions the stack as planned from inception — a Series-A team that mapped its credit needs upfront. The signal is small but it tips a meaningful percentage of partial-downgrade decisions toward full approval.

a real series-a engagement

VIIIAn anonymized Series-A $150K engagement — what the file looked like

The following is an anonymized snapshot of a Series-A B2B SaaS company's $150K credit engagement routed through CloudRoute in Q1 2026. Sensitive details (company name, exact lead investor, named individuals) have been redacted.

Company profile. Series-A B2B SaaS, 14 engineers, San Francisco. Round closed 9 months prior with a top-15 institutional lead and two known seed-stage co-investors retained from the previous round. Existing AWS account with $7K/month current spend, projected to scale to $11K/month over the credit window as the team rolled out a new compliance vertical and integrated AI summarization into the workflow.

Portfolio use case paragraph (anonymized). "We operate a B2B SaaS for revenue operations teams. Production infrastructure runs on ECS Fargate behind ALB, with Aurora Postgres for the operational database, S3 for object storage, and CloudFront for the marketing site. We also use CloudWatch + X-Ray for observability and Cognito for customer authentication. Projected monthly AWS spend over the next 18 months is $9K–$11K, scaling with customer count from the current 280 to a projected 480."

Build for Startups use case paragraph (anonymized). "We are launching a SOC 2 compliance vertical for enterprise customers, distinct from our core revenue-ops product. The vertical will use AWS Audit Manager, AWS Config, CloudTrail, and a dedicated Lambda-based evidence collection pipeline writing to a separate S3 bucket with stricter retention policies. Projected monthly spend on the vertical: $1.5K. 9-month implementation window, led by VP Engineering [redacted]."

Bedrock POC use case paragraph (anonymized). "We are evaluating Amazon Bedrock to power a customer-facing AI summarization feature for our revenue-ops dashboard. Chosen model: Claude Sonnet 4.5 for the cost/quality balance at our scope. Eval methodology: N=500 customer dashboard interactions per week, measured against a baseline rule-based summarizer, with manual review of disagreements scored against a held-out test set of 200 examples. Projected Bedrock inference spend: $1,400/month. 60-day POC window with go/no-go decision at Day 45."

Outcome. Three records filed Day 5. Portfolio approved Day 11 at full $100K. Build for Startups approved Day 13 at full $25K. Bedrock POC approved Day 17 at full $25K. Total stack: $150K. Total wall-clock from inquiry to all-balances-applied: 19 days. Total founder time across the engagement: 78 minutes. Total cost: $0.

What tipped this to full approval. The institutional lead carried tier-1 weight. The 9-month round age sat in the strongest signal window. The team size (14 engineers) sat at the upper end of Series-A engineering norms. The Portfolio projection ($9K–$11K) was defensible and supported by existing AWS Cost Explorer data the partner attached as evidence. The Build for Startups use case described a structurally distinct compliance vertical using a different AWS service surface (Audit Manager + Config + Lambda evidence pipeline) than Portfolio. The Bedrock POC plan included the four required components — concrete use case, named model with reasoning, eval methodology with sample size, and explicit budget with POC window. Each variable cleared the reviewer threshold independently. The compound result was full-ceiling approval on all three records.

side by side

Series-A pursuing $150K versus alternative target amounts at Series-A

Series-A teams sometimes ask: is the marginal effort to reach $150K worth it versus stopping at $100K (Portfolio only) or $125K (Portfolio + Build)? The honest math:

VariablePortfolio only ($100K)Portfolio + Build ($125K)Full $150K stack
Credit pools filed1 record2 records (same day)3 records (same day)
Founder time across application~30 min~45 min~70 min
Time-to-balance (full stack)11–18 days12–18 days18–21 days
Series-A full-ceiling approval rate~92%~85%~80%
Median outcome$100K (full)$120K (partial-Build common)$140K (partial-Bedrock common)
Required workload count1 (general infra)2 (infra + discrete project)3 (infra + project + AI)
Best fit at Series-ASingle coherent productProduct + distinct new workloadProduct + workload + AI/Bedrock POC
Marginal credits per additional founder hourN/A~$1,500/min~$1,000/min
The marginal credits per minute of additional founder time stay above $1,000/min through the full $150K stack. The constraint is whether the second and third workloads are genuine — adding fictional workloads to chase the ceiling triggers the overlap downgrade and produces worse outcomes than stopping at $100K honestly.
ready to file the series-a stack?
Get matched with a partner who files all three records same-day
Start in 3 minutes →
a recent series-a match

How the $150K stack lands for a Series-A AI company

inquiry · series-a ai-native b2b saas, austin
Series-A AI legal-tech

Situation: Series-A AI-native company building developer-tooling for back-office automation. $5M round 7 months ago, tier-1 institutional lead, existing AWS account with $6K/month current spend. Bedrock POC for the core inference workload was the primary credit target — the team had already migrated off OpenAI and was running Claude through Bedrock at scale. Wanted to maximize the credit stack to cover both general infrastructure scaling and the Bedrock POC layer for the production inference workload.

What CloudRoute did: Routed within 18 hours to a US-Central Premier-tier partner with Bedrock POC track record (filed 23 Bedrock POC ACE records in the prior 12 months). Partner ran the 45-minute discovery call on Day 2. Three records filed Day 5: Activate Portfolio ($100K base for general infrastructure), Build for Startups ($25K for a distinct internal-tools workload using Glue + Athena for the team's usage analytics product line), Bedrock POC ($25K with a Claude Sonnet 4.5 eval plan plus explicit production-scale inference budget).

Outcome: Portfolio approved Day 12 at full $100K. Build for Startups approved Day 14 at full $25K. Bedrock POC approved Day 18 at full $25K. Total stack: $150K. Total wall-clock from inquiry to all-balances-applied: 19 days. Total founder time across the engagement: 71 minutes. Total cost to customer: $0; CloudRoute commission paid by partner from AWS engagement funding.

engagement window: 19 days · founder time: ~71 min · credits secured: $150K · cost to customer: $0

faq

Common questions

Why is $150K specifically the modal target for Series-A and not $100K or $200K?
$100K is Portfolio alone — leaves $50K of stackable credits on the table when the team has a distinct second workload and an AI workload. $200K requires a different program family (Generative AI Accelerator, MAP) with materially different mechanics — competitive cohort applications, partner-led migration commitments, 60–180 day timelines. $150K sits at the highest credit-per-founder-hour ratio achievable through standard partner-filed Activate applications at Series-A, which is the band most Series-A teams should target.
Can a Series-A team without an AI workload still hit $150K?
No — the Bedrock POC layer requires a real AI use case with an evaluation plan. Without it, the realistic Series-A ceiling drops to $125K (Portfolio + Build for Startups). Trying to fake an AI workload to hit $150K is detectable — reviewers flag vague POC plans and downgrade to the $10K floor or reject outright. The honest target for non-AI Series-A teams is $125K via two pools.
What happens if my Series-A round closed 20 months ago — am I still in the strongest band?
Edge of the band. Reviewers will weigh the round-age signal more heavily and may ask whether a Series-B is in progress. The $150K stack is still achievable, but the partner should structure the application to include a clear path-to-next-round narrative or explicit traction signals. Approval rate at 18–24 months post-round is ~65% rather than the ~80% within 12 months.
Does the lead investor really matter to reviewers? It feels arbitrary.
It matters more than founders expect. AWS reviewers cross-reference the named lead investor against Crunchbase before approving full-ceiling applications. Tier-1 funds (top-15 globally by AUM and reputation) trigger immediate trust. Tier-2 funds trigger moderate trust. Solo capitals and angel syndicates require supplementary signals (engineering headcount, current AWS spend documentation, audit-stage posture) to clear the same approval bar.
Can I file the stack via my VC instead of an AWS partner?
If your VC is in the Portfolio Sub-Program and willing to submit within a week, yes — file via the VC. Same Portfolio ceiling ($100K), sometimes faster (10–14 days). The catch: the VC route typically does not extend to Build for Startups or Bedrock POC layers, both of which are partner-filed via ACE. Many Series-A teams file Portfolio via the VC and Build + Bedrock via a partner — a hybrid pattern that works when both parties coordinate.
What if Bedrock POC gets downgraded to $10K — does that ruin the stack?
No — it reduces the stack from $150K to $135K. The Portfolio and Build for Startups layers are independent of Bedrock POC and approve on their own merits. A $135K stack is still strong relative to alternatives. The fix for the Bedrock downgrade is the application content (clearer eval methodology, more specific use case, defensible inference budget) rather than the stack structure.
How does this compare to the Generative AI Accelerator for Series-A AI startups?
Generative AI Accelerator funds up to $1M but runs as a competitive cohort with ~5% acceptance globally and 60–90 day timelines. The $150K stack lands at ~80% approval rate within 21 days. Most Series-A AI startups file the $150K stack first (high-probability, fast) and apply for Gen-AI Accelerator separately for the larger pool (low-probability, slow). The two are not mutually exclusive — credits from one program don't block applications to the other.
Can I add a second Series-A round's credits later if my company raises another round?
No — Activate credits are per-company, not per-round. Once you've received your Portfolio ceiling, additional Activate credits aren't available at the next round. The path to more credits post-$150K is a different program family: MAP for a migration scope, Solution Provider Program for a billing discount, EDP for committed-spend pricing, or Generative AI Accelerator for the AI cohort track.
Will my AWS account team know about my credit balance?
Yes. AWS Sales and Solutions Architects assigned to your account can see the promotional credit balance and its expiration. This is occasionally useful — when the balance is nearing zero, the SA may proactively introduce you to the EDP team for committed-spend pricing. Most Series-A teams use this as the natural bridge from Activate credits at Series-A to EDP discounts at Series-B.

Get the full $150K Series-A stack in your AWS account in 21 days.

CloudRoute routes Series-A teams to AWS partners who file all three ACE records — Portfolio + Build for Startups + Bedrock POC — same-day, with the reviewer-trust signals that tip the application to full ceiling.

matched within< 24h
series-a approval rate~80%
time-to-balance18–21 days
cost to you$0
$150K AWS credits for Series-A startups — the canonical stack (2026) · CloudRoute