Blog · Page 8
Field notes.
Page 8 of 31. Browse the archive of RFP workflows, grounded-AI architecture, and proposal operations notes.
Three exec summaries I rewrote this week, part 10
The monthly exec-summary teardown, continuing. Three real openings rewritten this week. What was wrong, what I changed, and the rule each rewrite reinforced.
Clustering win themes across 200 past bids
How we cluster win-theme assertions across a corpus of past proposals to surface repeat themes, where the signal is real, and where the clustering is just noise dressed as insight.
Win-loss intelligence, Part 1 of 5: what to capture and when
The 18 fields we capture on every bid, why each one matters, and why most teams skip twelve of them. Part 1 of a five-part series on running win-loss as a habit, not a quarterly ritual.
The win-loss database schema, explained
The five tables behind PursuitAgent's win-loss intelligence feature: proposal, theme, outcome, debrief, and the join back to the RFP block that sourced the claim. With the SQL.
Win-loss intelligence starts on day one
Most teams collect the wrong signals after a bid, and the wrong signals compound. An opinion piece on what to capture from the moment the RFP lands, not the day after the award email.
The proposal-team staffing playbook (3, 15, 50 seats)
Proposal-team structure is volume-driven, not culture-driven. What the three-seat, fifteen-seat, and fifty-seat functions look like. Named roles, named failure modes, when to jump tiers.
A pgvector migration postmortem
An index rebuild that cost us 90 minutes of degraded search across a handful of tenants. What we changed in the runbook, and the piece of the migration we wish we had rehearsed.
The team we hired in year one, honestly
Roles we hired, roles we should have hired earlier, and one role we hired too early. A quick accounting of the first year of PursuitAgent headcount decisions, in the spirit of being useful to anyone about to make the same calls.
A full-year retrospective on shipping grounded AI
Twelve months of evidence on the grounded-AI thesis. The Stanford hallucination number measured against our corpus, four failure modes and which ones we closed, what changed under the hood, and what I would tell Q1 Bo.
Past-performance writing voice: past tense, specific, humble
Three real awarded past-performance narratives, anonymized. Why present-tense past performance reads like marketing, and why the humble voice wins more than the confident one.
Draft autocomplete latency, end to end
Typing lag, inference queue, streaming output. The three budgets that add up to the 240ms P95 we hold ourselves to, and what happens when any one of them slips.
DORA compliance showing up in DDQs
The EU's Digital Operational Resilience Act shows up in a visible fraction of recent DDQs. What the regulation asks for, what buyers actually want, and how to structure the response without inventing posture.
Prefer to see the product?
Take the 5-minute tour, or start a trial workspace and see PursuitAgent draft answers with citations.