Understand before building
Start with the problem, the people affected, and the constraints that shape a useful solution.
Software engineer · Product-minded builder
I'm Bolton Opondo, a software engineer combining product thinking, technical depth and engineering ownership to build reliable, impactful software.
Selected work
Product engineering on Microsoft Feed across web and mobile: shipping features, measuring engagement, and keeping the experience fast, accessible and reliable in production.
02 ProfessionalEvaluation metrics for information density and temporal relevance, offline baselines and experiments, and investigations into why AI-generated reports lose context or fall short.
03 ProfessionalCitation handling and rendering, monitoring and telemetry, and reliability investigations into how sources travel from an AI response to the screen.
04 IndependentEvaluating persistent memory for coding agents: a memory engine paired with an evaluation platform that asks whether past experience helps an agent on new tasks.
05 CommunityVolunteer website development for a community organisation, from AI-assisted early design exploration to guiding content and refining the site.
How I work
Start with the problem, the people affected, and the constraints that shape a useful solution.
Consider how a decision behaves in production, how it can fail, and how it can be maintained.
Stay accountable to the quality and usefulness of the work, not only the code that ships.
Look for evidence that a change improved the experience; be candid when the evidence is not there.
Research & experiments
A working question about the relationship between evaluation design, decision-making, and limitations.
Read research noteAbout
Over seven years of building production software, from product engineering on Microsoft Feed to the quality, evaluation and reliability of AI-powered experiences. I'm drawn to the point where software engineering, system design and AI research engineering meet.
More about meNow · updated 2 November 2026
Building: Memory Lab, an independent project that evaluates whether persistent memory helps coding agents on software engineering tasks.
Learning: Evaluation design, retrieval and grounding, and how to measure the quality of AI-powered experiences.
More about nowContact