About
Search is an engineering problem before it is a product problem.
StreamSearch is a small engineering practice focused on enterprise search: the indexes, pipelines and ranking layers that help organizations find and use their information. Alongside our engineering services, we are developing an enterprise search product for focused organizational pilots.
It exists because the pattern repeats. A company adopts a search engine, ships a first version, and then discovers that relevance, freshness and cost are three engineering disciplines nobody on the team has time to learn. Progress stalls somewhere between a working demo and a system the business trusts.
We come in at that point. Sometimes that means a two-week review and a roadmap the team executes alone. Sometimes it means months of building alongside your engineers. Either way the goal is the same: when we leave, the system is documented, measured and yours.
We work remotely with teams in Europe, the Gulf and North America, in English, and we are comfortable being the only search specialists in the room.
Built by Saeed Shahsavan
StreamSearch is built and led by Saeed Shahsavan, a software architect with 18+ years of experience in distributed systems, enterprise integration, event-driven platforms and search engine design. He holds a Master's degree in Software Engineering, with a thesis on the Design of an Enterprise Search Engine Based on Cloud Computing — the same problem StreamSearch is now built around. He is the author of Building Enterprise Projects with Go, and writes about search and distributed systems at saeedshahsavan.com.
How we work
Measure before you tune
No ranking change ships without an evaluation that can tell whether it helped.
Boring infrastructure wins
Proven engines and well-understood streaming beat novelty in systems that must stay up.
Hand over, do not hold on
Runbooks, architecture notes and pairing are deliverables, not afterthoughts.
Say no early
If the problem is not ours to solve, we say so in the first conversation.