Generative AIProduct
Shared memory is not a brief you would send
Claude's chat and Cowork now share one memory. I packed the same topic files into three agent tasks and checked which remembered asides left with the draft.
23 min readSplit
Shravan Pendem
From research paper to production to revenue.
I build AI products and work out who will pay for them. I started as an engineer at Samsung R&D, moved into strategy consulting at Accenture, founded an AI company that was acqui-hired in 2024, then joined Nurix AI as a founding member. Today I am a Research Project Manager at Deccan AI.
Those jobs taught me the same lesson from different angles. The hard part of enterprise AI is rarely the model. It is the distance between something that works in a demo and something that survives a real customer, a real budget, and a real security review.
Closing that distance is the work I care about, and it is what I write about here. Most essays take one research paper, rebuild its central idea, and ask a single question: would this survive production?
Writing
- 42 min
Ship the PaperLong-Horizon Planning & Reliability
A compaction verifier hears only the loud failures. The quiet ones decide the task
The paper scores a context summary by the blocked or repeated tool calls it causes over the next few steps. I rebuilt the scorer to ask how much of a summary's damage ever makes that noise.
Wait - 21 min
Generative AIInfrastructure
The model is not the assistant
I ran one scripted model inside four different apps and asked the same questions about one expense sheet. The app decided what came back right, and how the failures sounded.
Split - 20 min
Generative AIApplications
The prompt is not the application
I used to treat a good prompt as the whole application. Then I built the same launch email four ways and counted which numbers from the notes ended up in front of the customer.
Split
Selected work
Generative AI
Generative AI: Memory Is Not a Brief
The same Northwind topic files, packed six ways into three agent tasks. Live-update fixed the date. The speaker email still left with a brainstorm listed as a sponsor.
- Python
- Memory Packing
- Unittest
Ship the Paper
Ship the Paper: Context Compaction Verifier
An offline test of TRACE's compaction verifier, which scores a context summary by the tool errors it causes next. It removed the failures it could hear and none of the ones it could not.
- Python
- Closed-Loop Verifier
- Pass² Metrics
- Unittest
Generative AI
Generative AI: Model Is Not the Assistant
One scripted model inside four different apps, asked the same questions about one expense sheet. Two apps tied on score, with one failing loudly and the other confidently wrong.
- Python
- Harness Ablation
- Unittest
If you are working on enterprise AI and stuck somewhere between the demo and the contract, that is the conversation I enjoy most.