A Global Health Brand Team Shipped a BI Agent in 21 Days
~75%
2–4 months
Not disclosed
Brand teams at a multinational consumer health company, whose household-name hygiene, health, and nutrition products sell in more than 60 countries, could not get business intelligence answers without filing engineering tickets or waiting on analyst queues. The company's prior AI usage amounted to raw API requests to LLMs, leaving a large gap between ambition and practice. The screen the team most wanted to automate turned out to depend on nearly 20 separate data workflows.
Lazer structured the engagement as three one-week phases: discovery and data preparation, build, and iteration with feedback. The discovery week did the unglamorous work that makes agent projects succeed: mapping where the data actually lived, how it was queried, and which workflows fired when. Calling all 20 underlying workflows directly was not feasible, so Lazer simplified the SQL those workflows shared into a form an agent could use reliably.
On that foundation, Lazer built agent flows using the Retool Agents framework on the company's Databricks data layer: agents that automatically run workflows and surface insights for brand teams on demand. The engagement also shipped two enablement assets, an agent development guide and a POC agent-and-tools breakdown, documenting the gotchas so the client's own team could build the next agents themselves. The system went from kickoff to production in 21 days.
Lazer structured the engagement as three one-week phases: discovery and data preparation, build, then iteration with feedback. The first week focused on the unglamorous groundwork that makes agent projects succeed: mapping where the data actually lived, how it was queried, and which workflows fired when. The team found that the single screen brand users most wanted to automate depended on nearly 20 separate data workflows. Calling all of them directly was not feasible, so Lazer simplified the SQL those workflows shared into a form an agent could invoke reliably, without changing the numbers the business already trusted. On that foundation, the build week assembled agent flows using the Retool Agents framework on the company's existing Databricks data layer, creating agents that run workflows and surface insights to brand teams on demand. Claude supported prompt engineering during the build. Alongside the agents, Lazer produced two enablement assets, an agent development guide and a POC agent-and-tools breakdown, documenting the gotchas so the client's own team could build subsequent agents without further help.
Global consumer goods and CPG companies whose brand and marketing teams depend on analyst queues for routine data questions, and who want a working agent in weeks rather than a quarter-long platform program.
21 Days From Kickoff to Production
60+ Countries, 30+ Brands Self-Serving
~20 Workflows Collapsed Into One Agent Path
Aanikh Kler
Lazer ran a three-week engagement, one week of discovery and data prep, one week of build, and one week of iteration, so the client went from kickoff to a production agent in 21 days.
The agents were built on the Retool Agents framework running against the client's Databricks data layer, with simplified agent-callable SQL. Claude supported prompt engineering during the build.
Brand teams across 30+ brands in more than 60 countries now pull BI answers on demand instead of joining an analyst queue, and roughly 20 underlying workflows were collapsed into a single agent-callable path.
The build took 21 days from kickoff to production, split into three one-week phases: discovery and data preparation, build, and iteration with feedback.
It fits global consumer goods and CPG companies whose brand and marketing teams rely on analyst queues for routine data questions and want a working agent in weeks rather than a quarter-long platform program.
Have a similar challenge?
