Blog · Page 30
Field notes.
Page 30 of 31. Browse the archive of RFP workflows, grounded-AI architecture, and proposal operations notes.
Five years of APMP salary data, in one chart
What the APMP salary surveys tell us — and don't — about proposal-team compensation over the last five years. A teardown of the public dataset, the methodology, and the gaps a careful reader has to keep in mind.
In preview: per-block source permissions
Per-block permissions in the KB — control which teams or roles can read, draft from, or edit each content block, with audit trails on every read. In preview while the marketed KB surface catches up.
Stop CC-ing your SMEs
The 48% SME-bottleneck stat has held steady for five years. Most teams treat it as a people problem. It is a workflow problem. Replace the CC chain with a ticketed ask that has a deadline and a single owner.
Grounded Retrieval 101, Part 2: why citations don't guarantee groundedness
A citation tells you which passage was retrieved. It does not tell you whether the cited passage actually supports the generated claim. Part 2 of the Grounded Retrieval series — the entailment gap, and what closes it.
The compliance matrix you wish existed (Part 2 of 4)
The compliance matrix is the most consequential artifact in proposal work, and the one most teams build too late. Part 2 of Reading an RFP — what it is, how to build one in 30 minutes, and why it usually doesn't get built.
Embedding model selection: why Gemini Embedding 2 for proposals
A teardown of how we evaluated four embedding models — Gemini Embedding 2, OpenAI text-embedding-3-large, Cohere embed-v4, and Voyage — for a proposal corpus, and the methodology that drove the choice.
What the 2025 federal procurement data tells us about RFP volume
SAM.gov publishes the official record of federal solicitations and contract awards. Here is what the 2025 dataset says about RFP volume and what proposal teams should read into the trend.
Five patterns we see in Georgia state procurement postings
A structural read of patterns that recur across Georgia Procurement Registry solicitations. Five observations about how Georgia agencies draft RFPs and what each pattern tells a vendor about how the buyer thinks.
Shipped: content-block versioning in the KB
Every content block in your knowledge base now keeps an immutable version history. Drafts that cited a block stay tied to the exact text they cited, even after the block is edited.
The day we stopped saying 'AI' and started saying 'retrieval'
A short note on a vocabulary switch we made internally — and the reason a one-word change settled three months of recurring product debates.
Grounded Retrieval 101, Part 1: what RAG is and why it still hallucinates
RAG in three sentences, then the hard part: why retrieval-augmented generation still produces fabricated answers, and what the academic and practitioner literature says about it. Part 1 of a four-part series.
What 'reading an RFP' actually means (Part 1 of 4)
Reading an RFP isn't reading. It's six discrete passes — scope, compliance language, evaluation rubric, timeline and addenda, procurement signals, deal quality — each producing its own artifact. Part 1 of a four-part series.
Prefer to see the product?
Take the 5-minute tour, or start a trial workspace and see PursuitAgent draft answers with citations.