Launch HN: Sift Dev (YC W25) – AI-Powered Datadog Alternative
Hi HN! We're Kaushik and Ishir. We’re building SiftDev (https://app.trysift.dev/docs), an intelligent logging tool that understands your observability data in real time, automatically identifies anomalies, and lets you interact with your logs through natural language queries. Here’s a demo video: https://www.youtube.com/watch?v=uQ-TTdiu3fc&t=20s, and there's a demo playground you can try out here: https://app.trysift.dev/.
We used to work on product and engineering at Datadog and Splunk. We saw how even teams using these industry-leading tools were struggling to effectively interpret and use their logging data. The sheer volume of logs overwhelmed experts and newcomers alike, making it difficult to quickly identify meaningful issues or patterns. Despite powerful indexing and search capabilities, developers still had to manually piece together context from different logs, dashboards, and sources—a tedious and error-prone process.
The “noisy logging” problem—that is, the gap between overwhelming amounts of raw log data and insights people can act on—ultimately is a gap between machines (which generate all this data) and humans (who want and need the insights). SiftDev is built to bridge that gap and to automate the tedious, manual aspects of debugging and observability. In marketing-speak: “humans should never have to look at a log again!” We think people should interact with their data in terms that make sense on a human level.
What makes SiftDev different is its understanding of application context over time. While traditional tooling typically lets developers analyze logs in isolation, or with minimal surrounding context, SiftDev builds comprehensive profiles of your application's normal behavior patterns. This awareness allows us to understand what's truly abnormal versus what might appear unusual in a single snapshot but is actually expected behavior for your specific application. SiftDev applies semantic analysis and profiling to understand your application's logging behavior holistically. Instead of relying solely on manual search, Sift identifies core application processes, automatically detects patterns, and surfaces anomalies, including clear explanations and context.
Here are some examples of what this can look like in practice: Identify core processes: SiftDev instantly recognizes your payment workflows—like authorization, capture, and refunds—without manual tagging. Detect performance patterns: SiftDev learns your nightly batch job typically handles 10,000 records in 45 minutes, establishing a clear baseline. Surface hidden anomalies: SiftDev flags silent failures, such as two microservices updating the same record within 50ms—issues normally hidden by routine logs.
You can then directly ask your logs questions like, “What's causing errors in our checkout service?” or “Why did latency spike at 2 AM?” and immediately receive insightful, actionable answers that you’d otherwise manually be searching for.
We’d love for you to test out our product via our demo playground at https://app.trysift.dev/! It’s a slightly less functional version of our platform but shares a lot of the core features. Note: we do need users to sign up to do this but waitlist is optional (of course).
We'd love your feedback, thoughts, and experiences dealing with logging and observability challenges!
Can it run completely on prem ?
In most of the industries I work in we would never just send you our logs.
What stops me from building my own logger that sends a request to write a record to a DB and later asks an LLM what it means ?
Where is the pricing information?
Why do I need to login visit your homepage? How would I pitch this to my boss if they can’t read what it does ?
Edit: https://runsift.com/pricing.html
I see the landing page. The pricing should be clear though “ Contact Us” is scary.
> Can it run completely on prem ?
Yep we have an on-prem offering as well, got similar notes from folks before!
> What stops me from building my own logger that sends a request to write a record to a DB and later asks an LLM what it means ?
Great question! The main limitation over brute force is the sheer volume of noise, and therefore relevant context. We tried this and realized it wasn't working. From a numbers perspective, at even just 10s of GBs/day scale of data (not even close to enterprise scale), mainstream LLMs can't provide the context windows you need for more than a few minutes of operational data. And larger models suffer from other factors (like attention diffusion / dilution & drift).
> I see the landing page. The pricing should be clear though “ Contact Us” is scary. Noted!
Thanks!
I hope my tone wasn’t too brash.
If you can update the pricing I might be able to pitch this to my org later this year. We’d definitely like an on prem solution though!
Funny I was thinking this week logging needs some magic.
Log diving takes a lot of time especially during some kind of outage/downtime/bug where the whole team might be watching a screen share of someone diving into logs.
At the same time I am sceptical about "AI" especially if it is just an LLM stumbling around.
Understanding logs is probably the most brain intensive part of the job for me, more so than system design, project planning or coding.
This is because you need to know where the code is logging, imagine code paths in your head and you constantly see stuff that is a red herring or doesn't make sense.
I hope you can improve this space but it won't be easy!
Very relatable experience with log diving, feels very much like a needle-in-haystack problem that gets so much harder when you're not the only one who contributed to the source of errors (often the case).
As for the skepticism with LLMs stumbling around raw logs: it's super deserved. Even the developers who wrote the program often refer to larger app context when debugging, so it's not as easy as throwing a bunch of logs into an LLM. Plus, context window limits & the relative lack of "understanding" with increasingly larger contexts is troublesome.
We found it helped a lot to profile application logs over time. Think aggregation, but for individual flows rather than similar logs. By grouping and ordering flows together, it's bringing the context of thousands of (repetitive) logs down to the core flows. Much easier to find when things are out of the ordinary.
Still a lot of improvements in regards to false positives and variations in application flows.
The best way to improve this is to just generate decent useful and actionable logs. Sifting through a trash heap is where the problem is. No magic will suddenly turn that trash into gold.
You have to do this at the inception of the software you’re building rather then strap it on the donkey when something breaks (the usual way).
Yep, but it's sometimes a compromise people may be unwilling to make. Too often I hear (and have seen via DD customers) horror stories about initiatives to fix observability squashed by teams in hopes of shipping.
Moving fast has it's downsides and I can't say I blame people for deprioritizing good logging practices. But it does come back to bite...
Though as a caveat, you don't always have control over your logs -- especially with third party services, large but fragmented engineering organizations, etc. -- even with great internal practices, there's always something.
On another note, access to codebase + live logs gives room to develop better auto-instrumentation tooling. Though perhaps cursor could do a decent enough job at starting folks off
[flagged]
Disclaimer: I'm a founder at Gravwell, a log analytics startup
I agree, even when applicable LLMs are relegated to analyzing subselected data, so logs have to go somewhere else first. I think understanding logs is brain intensive because it can be a tricky problem. It gets easier with good tools, but often those tools are the kind that need to be used to build something else that solves the problem, rather than solve the problem themselves (e.g. building a good query + automation). I think LLMs can get better at creating the queries which would help a lot.
We started Gravwell to try bring some magic. It's a schema-on-read time-series data lake that will eat text or binary and comes in SaaS or self-hosted (on-prem). We built our backend from scratch to offer maximum flexibility in query. The search syntax looks like a linux command line, and kinda behaves like one too. Chain modules together to extract, filter, aggregate, enrich, etc. Automation system included. If you like Splunk, you should check us out.
There's a free community edition (personal or commercial use) for 2GB/day anon or 14GB/day w/ email. Tech docs are open at docs.gravwell.io.
> SiftDev flags silent failures, such as two microservices updating the same record within 50ms
I don't understand, what about that is a "silent failure"?
in order for your product to even know about it, wouldn't I need to write a log message for every single record update?
and if my architecture allows two microservices to update the same row in the same database...maybe it happening within 50ms is expected?
that could be an inefficient architecture for sure, but I'm confused as to whether your product is also trying to give me recommendations about "here's an architectural inefficiency we found based on feeding your logs to an LLM"
> You can then directly ask your logs questions like, “What's causing errors in our checkout service?” or “Why did latency spike at 2 AM?” and immediately receive insightful, actionable answers that you’d otherwise manually be searching for.
the general question I have with any product that's marketing itself as being "AI-powered" - how do hallucinations get resolved?
I already have human coworkers who will investigate some error or alert or performance problem, and come to an incorrect conclusion about the cause.
when that happens I can walk through their thought process and analysis chain with them and identify the gap that led them to the incorrect conclusion. often this is a useful signal that our system documentation needs to be updated, or log messages need to be clarified, or a dashboard should include a different metric, etc etc.
if I ask your product "what caused such-and-such outage" and the answer that comes back is incorrect, how do I "teach" it the correct answer?
> I don't understand, what about that is a "silent failure"?
Silent failures can be "allowed" behavior in your applications that aren't actually labeled as errors but can be irregular. Think race conditions, deadlocks, silent timeouts, or even just mislabeled error logs.
> in order for your product to even know about it, wouldn't I need to write a log message for every single record update?
That's right, and this may not always feasible (or necessary!), but if your application can be impacted by errors like these, perhaps it may be worth logging anyway.
> the general question I have with any product that's marketing itself as being "AI-powered" - how do hallucinations get resolved?
> and if my architecture allows two microservices to update the same row in the same database...maybe it happening within 50ms is expected?
> if I ask your product "what caused such-and-such outage" and the answer that comes back is incorrect, how do I "teach" it the correct answer?
For these concerns, human-in-loop feedback is our preliminary approach! We have our own internally running to account for changes and false errors, but having explanations from human input (even as simple as "Not an error" or "Missed error" buttons) is very helpful.
> when that happens I can walk through their thought process and analysis chain with them and identify the gap that led them to the incorrect conclusion. often this is a useful signal that our system documentation needs to be updated, or log messages need to be clarified, or a dashboard should include a different metric, etc etc.
Got it, I imagine it'll be very helpful for us to display our chain of thought from our dashboards too. Great feedback, thank you!
Your python sdk's <https://pypi.org/project/sift-dev-logger> GH link is 404: <https://github.com/sift-dev/python-sdk> Navigating upward shows the fork of SigNoz which I think is funny
There was no GH link for your npm dep so maybe they're both private. Although npmjs shows your npm one as ISC licensed, likely because of the default in package.json
Ah, any particular reason to want these SDKs public? Happy to, especially since you can see source on install anyway. Just curious!
And Kudos to SigNoz as well - have to check out other folks in the space :)
My initial concern was what transitive deps it was pulling in, but the other answer to your question is the thing that most GH repos are good for: submitting bugs and submitting fixes
It is also good for finding out what the buffering story is, because I would want to know if I'm dragging in an unbounded queue into my app (putting memory pressure on me) or knowing that your service returning 503s is going to eat logs. The kind of thing that only looking at the source would say for sure because the docs don't even hint at such operational concerns
Anyway, the only reason I mentioned the dead link is because your PyPI page linked to GH in the first place. So if you don't intend people to think there's supposed to be a repo, then I'd suggest removing the repo link
Noted, thank you! Will make some changes accordingly.
Neat idea. Why logs, and not metrics too? You can characterize an accurate "baseline" system behavior through a combination of system level and userspace metrics. This profile would offer more depth than what you'd otherwise piece together with userspace logs.
Agreed! Metrics are a high priority, especially since working to increase the available context around each anomaly we flag.
Logs were a natural starting point because that’s where developers often spend a significant amount of time stuck reading & searching for the right information, manually tracking down issues + jumping between logs across services. In a way, just finding & summarizing relevant logs for the user gave people an easier time debugging.
But metrics will introduce more dimensions to establish baseline behavior, so we're pretty excited about it too.
I tend to use logs the least when debugging production issues. I realize that's a personal anecdote, so I see your point.
TINLA, but perhaps you need to ensure your product complies with potential trademark issues related to sift[.]com.
Related https://www.datadoghq.com/product/platform/watchdog/
How does this compare with Axiom? I'm looking to shift out of Datadog asap and Axiom was the choice. Would consider Sift
Consider not marketing yourself as an X alternative when launching? That might fly in slide decks and investor meetings. But I don't know what Datadog is, and I certainly don't care, won't look into what DataDog is just so I can be qualified to learn about your product.
I guess it may be the case that you really know who your target is, but why miss the majority of the market and position yourself as pepsi on the same stroke?
Datadog is industry standard at this point, if you dont know what splunk or datadog is you are likely not their ICP and their marketing is not targeting you.
Agreed, if you don't know what Datadog is then you're probably not the target audience for this product.
Do you think if I don't know what datadog is, I am not the target audience for datadog?
probably
Hey - thanks for the feedback. We were trying to give people a good idea of where we fit in quickly, but I can see where you're coming from!
Even if it won't work for everyone — some people (including me) are looking for Datadog alternatives, so this is the easiest way for them to speak to their ICP.
Hi, I'm the author of LogLayer (https://loglayer.dev) for Typescript, which has integration with DataDog and competitors. Sift looks easy to integrate with since you have a TS library and the API is straightforward.
Would you like me to create a transport for it (I'm not implying I'd be charging to do this; it'd be free)?
The benefit of LogLayer is that they'd just use the loglayer library to make their log calls and it ships it to whatever transports they have defined for it. Better than having them manage two separate loggers (eg Sift and Pino for example) or write their own wrapper.
Hey, loglayer looks super cool! Would love to chat and set something up, send us an email at founders@runsift.com
Sent an e-mail!
curious how LLM hallucinations will work on logging info - gonna be a hard problem to solve
Java bindings would be welcomed by many.
Absolutely! Java bindings are on our radar. Any specific use cases / implementations you'd like to see? In the meantime, we do also support a couple off-the-shelf collectors that should already support Java applications!
What is your background to build 'AI powered datadog' alternate? Datadog is a massive company... how much experience do you guys have to have a product that competes with them?
OP literally said they worked at Datadog and Splunk. That’s enough tbh as those are leaders in this space
[stub for offtopicness]
Is it common practice to display fake realtime numbers on the homepage?
let storedNumber = getCookie("countingNumber"); let startNumber = storedNumber !== null ? storedNumber : Math.floor(Math.random() * (10300000 - 10000000 + 1)) + 10000000; let currentNumber = startNumber; function updateNumber() { let randomIncrement = Math.floor(Math.random() * (275 - 101 + 1)) + 101; currentNumber += randomIncrement; element.textContent = formatNumber(currentNumber); setCookie("countingNumber", currentNumber, 7); // Save number in cookie for 7 days } element.textContent = formatNumber(currentNumber); setInterval(updateNumber, 1000);
Ah! That was a leftover from the initial dev version of our website. I've taken it out now. Thank you!
The marketing design approach feels very off to me. You barrage me with an annoying scrolling marquee showing me the most abstract, unrecognizable logos telling me I should trust you because they do. 10+ companies on board feels rather small.
You said AI-driven analysis to identify logs, but I'm already skeptical of AI doing tasks like this, and you obfuscate it further by not actually showing me how it works, just another generic abstract marketing design graphic.
I dunno. It just seems like vaporware-as-a-service from the design vibes.
Early-stage startups often have websites that are little more than landing pages. That's because a full commercial website isn't in their critical path yet—first they need to build their product and attract early users, who don't typically come in through general web traffic.
That's one reason why Launch HNs usually include a demo video. That's the link you should be clicking on if you want to see these guys' product. If you do that, you'll see that it isn't vaporware.
We also advise startups doing Launch HNs to provide a link for users to try the product (preferably without a signup gate, but that's not always doable). There's such a link in the text above as well.
I suppose one way to avoid complaints about stub websites would be not to link to them at all—but then other comments would say "why would I trust you, you don't even have a website"!
Edit: I've replaced https://runsift.com/ with https://app.trysift.dev/docs in the text above. Perhaps that will help.
I want to jump in here and post this Launch HN form [0]. Obviously do not submit it if you are not a YC startup, but the questions on there are very helpful in terms of thinking about how to post about your startup on HN and elsewhere.
[0] https://docs.google.com/forms/d/1pRMkNiD-FKjYL-La5JWMwwrcWsp...
There's also https://news.ycombinator.com/yli.html, which is the guide for YC startups who want to launch on Hacker News. The formal mechanism is YC-only but the principles apply more broadly.
I'm not in YC, but I want to launch my startup here as it's relevant to the audience. Can I go through a process like this to coordinate with you for a launch, or should we just follow the guidelines, make a submission and hope for the best?
What's not recognizable about Duck, Square, Triangle, Asterisk, C, two different cubes, and the letter 'n'?
These, coupled with the random number generator to claim how many logs they're processing makes me wonder if the entire product is just AI generated slop.
Hey, maybe you can have a better hiring practice than datadog with a 5 question test where if you get a single answer wrong in even the smallest of ways you get disqualified from getting a job with them for 6 months.
I'm guessing they lost a wealth of great talent due to this test on how to support a platform that they give to fresh off the street applicants rather than having even a modicum of training about their product. They want you to study it for free, probably as a marketing tactic - but also so they don't have to pay to train employees. it's great like cancer.
Disclaimer: I have never applied to a role with datadog, nor interviewed with them. Just had multiple friends complete the process with mixed results. Seems like you need to put in ~two full weeks of self directed study to pass their on site interview 'exam' where they don't tell you about the exam being 100% or fail (but it is!)