Skip to content
Parsemend
Start free
roadmap

What's done, what's next, what we're only thinking about.

There are no dates on this page. A dated roadmap from a one-person company is a promise broken in public.

Shipped

Everything the rest of this site describes in the present tense.

  • Sentry-compatible ingest

    resolved

    Envelope and legacy store endpoints, all three DSN auth forms, unknown item types tolerated. Verified against the official PHP and Node SDKs.

    ingest

  • Grouping and triage

    resolved

    Custom fingerprints, in-app frame signatures, exception type and value, parameterized messages. Regression detection on the write path.

    grouping

  • Rate limits, quotas and retention

    resolved

    Per-key fixed-window limiting with X-Sentry-Rate-Limits. Events in a range-partitioned table, so expiring a day is a partition drop.

    quotas

  • Source maps and alert rules

    resolved

    Artifact upload with debug IDs, so minified frames resolve to your source. Alerts routed to email, Slack or a webhook.

    sourcemaps

  • The fix agent, in four modes

    resolved

    Context bundles, the resumable fix-run state machine, confidence gating, draft PR post-back. Bring your own agent, your own keys, the CLI, or managed.

    agent

  • Benchmark harness

    resolved

    Fix runs scored against a suite of real closed issues with known-good patches. Methodology and pass rate published, including the failures.

    benchmark

Next

Being built now. These are the two gaps we think matter most.

  • Sandboxed fix validation

    agent running

    Containers run your test suite against the agent's patch before the draft pull request is opened, and the result goes in the description. Seer never executes your code; this is the gap we most want to close.

    sandboxed-validation

  • Metric alerts

    agent running

    Alert on crash-free rate and error-rate anomalies rather than only on individual issues. Needs an aggregate layer that does not exist yet.

    metric-alerts

Exploring

Written down so nobody has to ask twice. Not commitments.

  • Session replay

    unresolved

    Sentry has it and we do not. It is a large amount of work for a feature the agent does not need, so it waits until someone tells us it is why they cannot switch.

    session-replay

  • Profiling

    unresolved

    Same reasoning as replay. Useful, expensive, and orthogonal to turning an error into a pull request.

    profiling

  • Tracing and performance

    unresolved

    Transaction envelopes are already accepted and dropped. Storing and querying them is a different database problem than storing errors.

    tracing

  • Analytics query layer

    unresolved

    Arbitrary queries over event attributes, which today means a column store we have deliberately not adopted.

    analytics

Three of the four items under Exploring are things Sentry already has. If replay, profiling or tracing is why you cannot switch, say so — that is the signal that moves them up, and nothing else will.