Recent reports suggest China has completed a prototype extreme ultraviolet (EUV) lithography machine — a milestone long thought years away — as part of a concentrated state-led effort in Shenzhen. The system reportedly generates EUV light but hasn’t yet produced working chips, and advanced lithography remains dominated by ASML’s commercial machines.
We’re building debugging intelligence for modern software teams.
BetterBugs captures rich technical context at the moment a bug occurs and turns it into actionable insights for developers — not screenshots, not back-and-forth.
Our next leap: AI that understands bugs, not just describes them.
Senior AI / LLM Engineer (Python) : Build production LLM systems that understand real-world bug context (logs, stack traces, network, user actions).
Design agent workflows for root-cause analysis, bug triaging, and debugging assistance.
Implement RAG pipelines grounded in codebases, historical bugs, and execution data.
Own prompting, evaluation, reliability, and observability of AI outputs.
Build and operate Python services (FastAPI) powering AI features in production.
Strong Python, hands-on LLM/agent frameworks, Docker; bonus for dev tooling or observability experience.
Thanks for the suggestion — I’m happy to schedule a short intro call.
I’ve selected *13:00* as the preferred time.
For context, I’m a senior backend/systems engineer with strong Python experience, currently focusing on building production LLM-powered systems for debugging, incident analysis, and developer tooling. I’m particularly interested in BetterBugs’ approach to understanding bugs through real execution context (logs, traces, user actions), rather than surface-level descriptions.
Looking forward to the conversation and learning more about your vision and current needs.
AI will kill flaky UI scripts, and “click-and-verify” roles. That’s overdue. What won’t disappear is the need to understand how systems actually behave under stress, failure, and ambiguity.
The future of QA is upstream:
• defining invariants, not writing scenarios
• modeling state and failure modes, not chasing bugs
• debugging distributed, async, messy real-world systems
AI will generate tests faster than humans ever can. But it won’t know what matters or what assumption is wrong. That judgment still belongs to engineers.
If you’re in QA and want to stay relevant: stop being a test executor. Become the person who explains why the system broke, not just how to reproduce it.
Upstream means Shiftleft in better words, we follow from while back , but yet not get solid success to automate mundane flows,
Using Playwright MCP a lot , but always worried about unknown unknown zone before every release.
I agree with the core point: the more pure and deterministic a system is, the easier it is to reason about and test. Reducers + property-based testing push correctness into design, not brittle test cases.
One nuance though: property-based testing shines when the domain is already well-modeled. A lot of real QA pain comes where purity breaks down—distributed systems, async flows, partial failures, UI↔backend boundaries. At that point, the hard part isn’t generating tests, it’s reconstructing context.
On LLMs: I don’t think they should be trusted as correctness oracles either. Their real value isn’t guessing answers, but helping surface assumptions, generate counter-examples, and expose gaps in our mental model.
So the future of QA isn’t humans vs LLMs. It’s better system design + explicit invariants + tools that help engineers doubt their own certainty faster. Most serious bugs come from being sure we understood the system when we didn’t.
The LIBR logic is straightforward, but OCR quality, auditability, and evidence integrity are what make this usable in the real world.
lawyers care about chain of custody, auditability, and immutability makes this less of an “AI app” and more of a compliance workflow tool, which might matter a lot for positioning.
On B2C vs B2B: individuals feel this once, lawyers feel it every case — which usually determines who actually pays.
The biggest risk seems less about accuracy and more about how courts classify the output (calculator vs expert opinion). That likely drives both liability and pricing.
Have you run this past a practicing family lawyer or forensic accountant yet, even informally?
OP here – thanks for the feedback. I just pushed an update to address the Chain of Custody concerns.
The system now generates immutable forensic reports with SHA-256 integrity hashes for every document. Also added a regression suite to verify the tracing algorithm against known edge cases. The focus is definitely shifting from just "AI wrapper" to "Audit Compliance tool.
I’m cautiously optimistic about AI, but less about the hype cycle around it.
AI (especially LLMs) will likely stay top-of-mind in 2026, but I expect costs to drop meaningfully and capabilities to feel more “infrastructure-like” rather than magical. SME Adoption will drive the AI to masses.
If AI doesn’t meet near-term revenue and productivity promises, we may see pressure or stagnation in tech valuations, even as the underlying technology continues to improve. In other words, the market may cool before the tech does.
On the macro side, I wouldn’t be surprised if we see more market stability or mild declines, driven by a re-rating of expectations rather than a systemic collapse. Capital might rotate from speculative growth into cash-flow-positive businesses that actually deploy AI profitably.
More broadly, I think 2026 will reward:
reliability over flashy innovation in AI , Engineering depth over marketing narratives, Systems thinking over isolated “features”
Less “what’s possible?” and more “what actually works at scale?”
Perhaps. Sometimes the scale is "one" - the amount of engineering that goes into bespoke space missions is very large, and very little of that work is re-used for anything other than direct follow up missions
Whole PLG motion depends on this.