CARR Biosystems × BioCreative Strategies
Full timeline

§01 — The Build

A four-month build.
A live system.
Yours.

Three months of foundation. One month of activation. The deliverables, the platform, and the messages going out — all yours. None of the customer-facing work is locked behind a vendor contract.

DELIVERABLE 01

Brand & positioning system

4 master docs · 17K-char positioning JSON · 30 research documents

DELIVERABLE 02

Universe + ICP v2 scoring

3,964 companies, 102K contacts, 7-tier taxonomy

DELIVERABLE 03

SSO matrix + AI message engine

37-row matrix, per-rep voice rules, contextual angle selection

DELIVERABLE 04

Outreach platform — both channels

LinkedIn (HeyReach) + Email (EmailBison) live

DELIVERABLE 05

Reply loop + DNC compliance

Webhooks → drafts → rep review → send

DELIVERABLE 06

3 marketing landing pages

Autologous · Accelerate · Scale-Up — on CARR brand

§02 — The Engagement

Four months. Four shifts.

Each month had a center of gravity. Below is what actually got built and shipped — and where Month 4 is going.

Month 1

Foundation

  • Platform stood up — full Supabase + UX deployed
  • 3,117 companies + 75K contacts loaded
  • 20-page UX shipped
  • ICP v1 scoring across 2,323 accounts
  • 5 LinkedIn senders + 6 campaigns live
  • 3 email domains warming

Month 2

The Messaging Engine

  • Brand merge: 44-page site scrape → 4 master docs
  • 37-row SSO messaging matrix
  • AI message generator (Claude Sonnet 4)
  • 7-persona contact classifier
  • Territory system across 6 reps
  • 3 marketing landing pages live

Month 3

Hardening

  • ICP v2 scoring rebuild (7-tier taxonomy)
  • 3-wave security audit
  • Per-rep writing rules + DNC compliance
  • Rep-note validation pipeline (518 verdicts)
  • 4 email campaigns staged · 500 contacts loaded
  • 446 migrations · 18 edge functions · 95 views

Month 4 · Current

Activation

  • Email campaigns live across 4 reps
  • A/B variant testing in production
  • Continual learning from real reply data
  • Newsletter pilot — CARR Market Pulse
  • BW Azure migration scoping underway

Worth noting

Send/receive automation was never in the original SOW. The Launch program scoped strategy + matrix + content. The reply loop, dual-channel sending, webhook reconciliation, and per-rep DNC enforcement got built anyway because the system needed them to function end-to-end.

§03 — Foundation

One unified target universe.

3,964 companies. 102,312 contacts. Every list, every spreadsheet, every "I have a contact at…" landed in one structured place — refreshed on a known cadence.

CONSOLIDATED FROM

  • Existing CARR account list + historical deals
  • Apollo + Clay enrichment passes
  • Conference + trade-show captures (39 events tracked)
  • Inbound + rep-sourced contacts

LIVE DATA SIGNALS

  • 1,862 clinical trials matched to accounts
  • 5,000 NIH grants matched to accounts
  • 704 curated news articles (refreshed daily)
  • Patents + regulatory filings on tracked targets

§04 — Enrichment + ICP

Three enrichment layers. One ICP score.

A name and a domain isn't a target. The system runs three enrichment passes per company, then derives a deterministic score. Apollo and ZoomInfo don't carry "what separation tech does this CGT company use" — because that requires CARR domain reasoning.

LAYER 1 · HUB ENRICHMENT — UNIVERSE-WIDE

Rule-based + Claude AI deep research

Every CGT company in the BioCreative universe gets the same baseline treatment. L1 produces structured subcategory + life-sciences flag. L2 runs Claude Sonnet over web content + scraped sites and writes a 22-key JSONB.

modality clinical_phase funding_stage technology_platform pipeline_products total_funding key_investors competitors ai_summary

Coverage: 75% of CARR's universe (gaps are mostly very-small + stealth-mode companies)

LAYER 2 · CARR-SPECIFIC — CUSTOM

Brave Search + Firecrawl + Claude → carr_enrichment_data

Custom code that takes the company's domain, web-searches it, scrapes the company's own site (about pages, technology pages, news), and runs Claude with CARR-specific extraction prompts. Output is purpose-built for the centrifuge-fit conversation.

centrifugation_relevance manufacturing_scale bioreactor_type separation_technology key_partnerships equipment_signals carr_fit_summary competitive_overlap objections_anticipated

This doesn't exist off-the-shelf — it requires CARR domain knowledge encoded in the prompts + schema

LAYER 3 · ICP V2 SCORING — DETERMINISTIC

calculate_icp_score_v2() — PL/pgSQL function

Reads both enrichment layers + live signals (trials, grants, news), runs a 4-dimension score, assigns one of 7 account tiers. No LLM in the scoring step — same inputs always produce the same tier.

30 PTS

Technical Fit

25 PTS

Market Opportunity

25 PTS

Competitive Position

20 PTS

Right to Win

7-tier taxonomy: cold · nurture · validated_target · customer · disqualified · parked · unreviewed

Where it lands

All three layers land as structured data on the Account page — a Gold Sheet view that surfaces everything in one place. The contextual engine reads from this directly when generating a message. Reps can flag bad data through a validation pipeline (518 rep verdicts to date) that re-checks claims against fresh web searches.

§05 — The People

From contact list to buying persona.

A title alone doesn't tell a rep how to talk to someone. Four passes, structured outputs, then a runtime priority chain that resolves a buying persona for every contact — even when classification hasn't run yet.

01

Discovery

Name, title, LinkedIn URL, sourcing context. Title parsing extracts structured seniority + department, normalized in a Postgres trigger on every insert.

02

Clay enrichment

Waterfall pulls the LinkedIn headline, About summary, experience arc, and best-confidence email. This step gates the next one — only Clay-enriched contacts proceed to classification.

03

Claude classifier → 6 marketing personas

Claude reads Clay output + company context, writes one of 6 CARR-specific personas plus a rich segment_data JSONB (persona summary, messaging angle, pain points, value hook, decision role).

executive_leadership technical_leadership research_leadership operations_leadership quality_regulatory_leadership clinical_leadership
04

Buying-persona derivation at runtime

A 6-step priority chain resolves a buying persona for every contact: buyer_segmentmarketing_personarole_typedepartment → seniority → executive default. Every contact gets a persona — quality degrades down the chain, but messages still get generated.

The contextual engineering layer

By the time the SSO matrix runs, every contact already carries enrichment + persona + buying angle + pain points + value hook. The LLM never has to guess.

§06 — The SSO Matrix

A context engineering agent — not a prompt.

When it's time to write an email or LinkedIn message, we don't ask an LLM to "be creative." We run the SSO Matrix — pull everything relevant about this person, this account, this moment, assemble it into a structured payload, and hand it to a static LLM for generation.

1

Pull

Enrichment record (Hub L2 + carr_enrichment_data) · ICP tier · persona segment_data · recent triggers (trials, grants, news) · brand voice tokens · message thread history.

2

Assemble

Context engineered into the SSO payload. The 37-row messaging matrix maps persona × ICP group × messaging variant → single sales objective + value props + lead-in phrasing. Per-rep voice rules attach as guardrails.

3

Generate

Claude Sonnet 4 (gpt-4o-mini fallback) writes the message. At runtime, not from a template. Output captures which tier of the matrix fired (db_matrix · db_matrix_cgt · role_fallback).

4

Score + ship

Brand Voice Guardian + per-rep writing rules + DNC check → send via HeyReach (LinkedIn) or EmailBison (email). Drafts surface in the rep's review queue first; nothing autonomous.

Co-authored, not vendor-written

Nico contributed his voice rules directly — three banned phrases, two required, structural rules — codified end-to-end. Jake contributed LinkedIn-specific patterns. Every cell of the matrix was reviewed by a CARR rep before it shipped.

Following Karpathy's method

The LLM is a static processor. The intelligence comes from the context engineering — what we pull, what we filter, what we hand it. All the work happens before the model sees the task. That's why every message reads like it was written by someone who knows the recipient's business.

§07 — The Platform

Where the work actually happens.

Four core surfaces, each one designed around what a rep actually does — not around how the database is structured.

[ Account Detail · Gold Sheet ]

01 · ACCOUNT DETAIL

Gold Sheet view

Miller-Heiman framing made operational — decision-makers, blockers, current/desired state visible per account. All three enrichment layers + signals consolidated on one page. Live pipeline tracking with deal stages sits on the Month 4+ menu.

[ Target Universe + List Builder ]

02 · TARGET UNIVERSE + LIST BUILDER

Filter → save view → push to campaign

Multi-axis filters (modality, stage, ICP tier, manufacturing scale, separation tech, recent grant, last contacted). Save the result as a list, push that list straight into a campaign.

[ Campaign Hub + Outreach Modal ]

03 · CAMPAIGN HUB + OUTREACH MODAL

Generate / review / send on one screen

Pull up a contact, see the SSO-generated message, edit if needed, send through HeyReach or EmailBison. No tab-switching to external tools.

[ Validation Queue + Data Quality ]

04 · VALIDATION QUEUE

Human review for edge cases

Anything ambiguous — bad email, conflicting persona, low-confidence ICP, rep-flagged data — lands in a queue, not a black hole. A rep reviews and re-routes; 518 verdicts logged to date.

§08 — Real Messages

What's actually going out.

Pulled live from the database. Six emails sent in the last week of May. Two LinkedIn DMs from active sequences. These are real messages, real recipients — light edits only for layout.

Bethanne Deuel → Gloria Zhang · President, AbVision Inc
Subject: AbVision's cell therapy pipeline and manufacturing readiness Hi Gloria, AbVision's approach to harnessing antibody discoveries for immune-based cancer therapeutics is compelling, especially as you advance toward clinical development. The transition from preclinical to Phase I often reveals manufacturing bottlenecks that weren't apparent at smaller scales. Many cell therapy developers we work with face viability loss during scale-up, particularly when transitioning from research-grade to clinical manufacturing processes. At Carr, our tubular bowl centrifugation platform consistently delivers >95% cell viability across CAR-T and other immune cell types, maintaining the critical quality attributes that determine therapeutic efficacy. Our applications team has supported several Bay Area biotechs through this exact transition, helping establish scalable processes early to avoid costly tech transfer delays later in development. Would you have 15-20 minutes next week to discuss how manufacturing strategy might impact your clinical timeline? Best regards, Bethanne Deuel CARR Biosystems
Bethanne Deuel → Betty Dean · Project Manager, Avobis Bio
Subject: Avobis Bio's MSC programs and processing considerations Hi Betty, Avobis Bio's approach to combining mesenchymal stem cells with bioabsorbable scaffolds represents a compelling strategy for regenerative healing. As your programs advance through clinical development, maintaining consistent MSC quality becomes increasingly critical under regulatory scrutiny. Many teams we work with face challenges preserving MSC viability during harvest and concentration steps, especially when scaling protocols for higher clinical volumes. At Carr, our tubular bowl centrifugation technology consistently delivers >95% cell viability while minimizing the shear stress that can compromise MSC therapeutic properties. Our applications team has supported several MSC-focused companies through similar scale-up transitions, helping them establish robust processing workflows that maintain cell quality attributes from development through commercial manufacturing. Would you have 15-20 minutes next week to discuss how other teams have addressed MSC processing challenges during clinical advancement? Best regards, Bethanne Deuel CARR Biosystems
Nico Summaria → Dan Yerace · Co-founder + VP Operations, COEPTIS
Subject: Scaling CAR-NK manufacturing at Coeptis Hi Dan, Coeptis's CAR-NK platform represents a compelling approach to solid tumor targeting, especially as you advance through clinical development. The challenge many teams face at your stage is ensuring the manufacturing processes that worked in early development can scale seamlessly to support later-phase trials without compromising cell quality. At Carr, our applications team regularly works with CAR-NK developers who need to maintain >95% viability while scaling from bench to GMP volumes. Our closed-system platform consolidates harvest, wash, and concentration steps that traditionally require multiple pieces of equipment and extensive manual intervention. We've seen similar clinical-stage companies reduce their tech transfer timelines significantly by implementing scalable bioprocessing early, before GMP requirements lock in suboptimal workflows. Would you have 15-20 minutes next week to discuss how other CAR-NK teams have approached scalable cell processing? Best regards, Nico Summaria CARR Biosystems
Nico Summaria → Dorit Harati · Sr VP Quality & Regulatory, NeuroGenesis Bio
Subject: NG01 manufacturing considerations for pivotal trials Hi Dorit, NeuroGenesis's NG01 platform for transforming marrow-derived cells into neuroprotective bio-factories represents a compelling approach to progressive MS and ALS. As you advance through clinical development, maintaining patient-specific batch integrity becomes increasingly critical under regulatory scrutiny. Autologous manufacturing introduces unique challenges around cross-contamination prevention and per-patient traceability that traditional processing equipment wasn't designed to handle. At Carr, our single-use tubular bowl centrifugation maintains >95% viability while providing complete lot documentation and gamma-irradiation records for each patient batch. Our applications team has worked extensively with clinical-stage cell therapy companies navigating similar autologous manufacturing requirements, particularly around 21 CFR Part 11 compliant data logging and batch-to-batch consistency. Would you have 15-20 minutes next week to discuss how automated, closed-system processing could support NG01's path through pivotal trials? Best regards, Nico Summaria CARR Biosystems
Jeff Plambeck → Andrew VanVurst · COO + Co-founder, BioStem Technologies
Subject: BioStem's BioRetain processing and scaling considerations Hi Andrew, BioStem's focus on preserving tissue structure and growth factors through the BioRetain method represents a thoughtful approach to perinatal tissue processing. As you scale manufacturing operations for your regenerative therapy portfolio, maintaining those critical quality attributes becomes even more challenging. Many tissue banking operations we work with face viability losses during scale-up when moving from manual processing to larger volumes. At Carr, our closed-system centrifugation platform has helped similar companies maintain >95% cell viability while scaling from 500 mL development runs to 2,000 L commercial batches without technology changes. Our applications team has worked extensively with regenerative medicine companies navigating the transition from research-grade processing to GMP manufacturing, particularly around maintaining consistent concentration factors and cell quality across scales. Would you have 15-20 minutes next week to discuss how other tissue banking operations have addressed scaling challenges while preserving their proprietary processing advantages? Best regards, Jeff Plambeck CARR Biosystems
Jeff Plambeck → Lyndsey Crump · Senior Scientist, TrAMPoline Pharma
Subject: MightyTIL scale-up considerations Hi Lyndsey, TrAMPoline's engineered TIL approach with enhanced anti-tumor activity sounds promising, especially with the preclinical data you're generating. As you move toward IND-enabling studies, the transition from discovery-scale TIL expansion to GMP-ready processes often creates unexpected bottlenecks. Many teams we work with hit viability drops during scale-up when switching separation technologies between development and manufacturing. At Carr, our tubular bowl centrifugation maintains >95% TIL viability while providing the throughput needed for patient doses, using the same low-shear technology from lab through commercial scale. Our applications team has been working with several TIL developers on process design early in development to avoid costly tech transfer issues later. We've seen this approach save months in IND timelines. Would you have 20 minutes next week to discuss how other TIL companies have approached scalable bioprocessing workflows? Best regards, Jeff Plambeck CARR Biosystems

§09 — The Reply Loop

When someone writes back, the system catches it.

A connection accept, a reply email, a meeting request — all routed back into the platform automatically. The rep gets a draft response, edits, sends. Nothing falls through because someone forgot to check an inbox.

1

Webhook capture

HeyReach + EmailBison post events directly to the platform — connections, replies, bounces, opens.

2

Match to contact + thread

Every event resolves to the right contact, the right campaign, the right historical thread.

3

Generate draft response

The SSO matrix runs again, this time with the reply text + thread history as context. Output: a starting draft.

4

Rep review

The draft surfaces in the rep's queue. They edit, approve, or reject. No autonomous sending of replies.

5

Send + log

Goes back through the same channel it came in on. The thread updates. The next message will see this as context.

Beyond the original SOW

The Launch program scoped strategy, matrix, and content. The reply loop got built because without it, every message turned into a guess about whether the recipient ever responded.

Recent fix (May 31): HeyReach webhook v2.6 — corrected a 99-day silent-failure bug in payload normalization. The loop is now confirmed reliable end-to-end.

§10 — Marketing Support

Three landing pages. One reusable engine.

Built on the CARR brand system. Live on the Hostinger VPS. Ready as outreach destinations once UTM tracking is wired.

The engine, not just the output

These three pages came out of a templated landing-page system that can also produce: event-specific pages (ASGCT, ISCT booth follow-ups), sales collateral (one-pagers per persona × modality), internal project briefs, and SEO landers targeted at long-tail manufacturing queries.

§11 — Coaching & Learning

A system the team is learning to operate.

Day-one philosophy: this is yours to run. Every layer of the build had a CARR co-author and a working session, not a hand-off.

Jake — parallel build + Claude Desktop

Owns the front-edge of the LinkedIn channel. Connected to the CARR DB via Claude Desktop + GitHub MCP for independent database access. Co-built the SSO matrix from the rep side.

Nico — voice-rule co-author

Three banned phrases, two required, structural rules — all his. The system runs them every time it generates under his name. Notes audit (April 22) drove the rep-validation pipeline.

Bethanne, Jeff, Andrew — operating reps

Each running their own active campaigns in Month 4. Rep-driven message review, daily flow through the platform. Each has their own voice profile in production.

Super-user cohort framing

The natural Month 4+ extension: a structured group inside CARR that doesn't just use the system but extends it — building new lists, new agents, new workflows on top.

§12 — Month 4+ Options

Many ways the platform can grow from here.

Through strategic education and support, additional build-outs, market segmentation, and integrations. A menu, not a contract — pick one, pick three, pick none and just keep refining what's running.

EDUCATION + SUPPORT

Continue & refine

Keep both channels running. Iterate on the SSO matrix from real reply data. Tighten ICP v2. Restore weekly strategy cadence. The compound-interest path.

EDUCATION + SUPPORT

AI super-user cohort + agent build

A 4–6 person group inside CARR trained to build and own custom agents on top of the platform — research agents, follow-up agents, internal-Q&A agents. Methodology + working time.

BUILD-OUT

Live Miller-Heiman pipeline

Take the Gold Sheet beyond static — every account in active opportunity scored on MH dimensions, refreshed off real activity. Pipeline reviews become data, not memory.

BUILD-OUT

Marketing collateral engine

Same templating system that built the 3 landers, repurposed for one-pagers, persona briefs, conference handouts, internal project pages. On-brand, fast, repeatable.

BUILD-OUT

Events workstream

Pre-event target lists, on-event messaging, post-event reply loop — all wired into the same platform. ISCT, ASGCT, PEGS, BIO each get a structured campaign treatment.

MARKET SEGMENT

Biologics line expansion

A second ICP, a second SSO matrix branch, a second set of landing pages. Same engine, different product line. Run both pipelines side by side.

INTEGRATION

HubSpot bridge

Two-way sync between the platform and HubSpot. Marketing keeps HS, sales gets the CARR platform — both stay current without copy-paste.

INTEGRATION

Azure migration (BW IT)

Three migration paths: keep on current infra, lift-and-shift to Azure, or selective workload move. We can scope and own whichever route the BW IT conversation lands on.

CHANNEL

SEO & organic content

Keyword clusters, pillar pages, Schema.org markup, and backlink strategy targeting bioprocessing decision-makers. Build domain authority so CARR ranks for high-intent commercial queries.

CHANNEL

Paid digital (LinkedIn + Google)

Precision-targeted LinkedIn campaigns by persona and account list. Google Ads on high-intent keywords. Retargeting via pixel on the landing pages already deployed.

CHANNEL

Outbound social

Branded presence on LinkedIn — company posts, rep amplification sequences, thought-leadership content calendar. Systematic not sporadic.

CHANNEL

Social listening & intent signals

Monitor bioprocessing conversations, job postings, expansion signals, and competitor mentions. Feed real-time buying intent back into the SSO matrix for timing-aware outreach.

BUILD-OUT

Full AI operating system

The complete agent architecture — coordination agents managing campaign flows, conductor agents orchestrating multi-step research, director agents making prioritization decisions. A self-improving system that runs outreach, monitors results, and adapts without manual intervention.

A menu, not a contract

The platform is built to take any combination of these without rework — that was the whole point of the foundation phase.