[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]

@0xlelouch_ Avatar @0xlelouch_ Abhishek Singh

Abhishek Singh posts on X about redis, if you, clickhouse, at a the most. They currently have XXXXX followers and XXX posts still getting attention that total XXXXX engagements in the last XX hours.

Engagements: XXXXX #

Engagements Line Chart

Mentions: XX #

Mentions Line Chart

Followers: XXXXX #

Followers Line Chart

CreatorRank: XXXXXXX #

CreatorRank Line Chart

Social Influence

Social category influence technology brands stocks finance social networks

Social topic influence redis #12, if you, clickhouse #3, at a, to the, s3, target, flow, breakdown, for all

Top accounts mentioned or mentioned by @grok @hyperliquidx @watcherguru @0xlelouch @ryan_improvises @abidsensibull @justinskycak @0xffdevs @bryanjohnson @chshersh @getpochi @yashkrverma @pandarocks2020 @vladchatware @thomasburkhartb @rachmaninoffian @suinetwork @stacksorg @gkcs @interviewready3

Top assets mentioned IBM (IBM) Microsoft Corp. (MSFT)

Top Social Posts

Top posts by engagements in the last XX hours

"We obsess over the final target (application logic) but the request has already lost XX% of its speed before it even reaches our code. Heres the fact: Most latency is created before your handler runs. Not inside it. Every hop adds drag: - TLS handshakes - Load balancer queues - Reverse proxies - NAT translation - Kernel networking stack - Thread scheduling - Memory allocation - Serialization/deserialization - JSON parsing - ORM magic - Logging hooks - Middleware pipelines By the time your app touches the request its already a tired arrow. Thats why the fastest systems arent the ones with the"
X Link 2025-12-04T15:58Z 4318 followers, 5233 engagements

"A lot of young people dont struggle because they lack talent. They struggle because no one ever showed them what talent actually looks like in the early stages (speaking from personal experiences). Real talent development is ugly. It looks like confusion boredom repetition slow progress self doubt and the quiet fear that maybe youre not built for this. Most kids think theyre failing because no one told them that this is exactly what the process is supposed to feel like. When you show a young person the real arc - how skill compounds slowly how bad drafts become good ones how repetition"
X Link 2025-11-30T22:32Z 4322 followers, 81.5K engagements

"Great breakdown same story repeats itself with Datadog just with a different shape of pain. Datadog becomes expensive not because its bad but because: Most teams turn on everything all at once without guardrails. Where Datadog gets you: X. Metrics cardinality explosion Developers add one innocuous tag (user_id request_id or worse session_id) suddenly a single metric becomes millions of time series retention blows up queries slow down costs skyrocket. Cardinality kills teams long before ingestion does. X. Logs stored in Datadog instead of forwarded to cheaper storage Teams often keep logs in"
X Link 2025-12-04T08:03Z 4331 followers, 1505 engagements

"1. When you have high reads and writes (balanced workloads): - Use a primary database for all writes. - Offload heavy read traffic to multiple read replicas via async replication. - Use Redis cache to handle hot keys and reduce database load. - Expect occasional cache misses to fall back to the primary. - Batch writes and optimize indexes to keep performance stable. X. When your write load keeps growing: - Shard the database. - Each shard stores a slice of the dataset so writes get distributed. - Your app must know how to route requests to the right shard. - Great for massive scale tricky for"
X Link 2025-12-01T08:04Z 4387 followers, 14.4K engagements

"I have been working with kafka recently for writing a consumer in golang that simply reads data from kafka and saves it to clickhouse for audit purposes. Encountered a bug while migrating the consumer from confluent to AWS MSK (managed serverless kafka). The connection was reset by peer from time to time but this happened only in case of AWS MSK and not confluent kafka. Debugged the issue this happens because AWS MSK traffic normally flows through an AWS Network Load Balancer whose default idle timeout is 350s. If no I/O happens before that deadline the NLB silently closes the socket yielding"
X Link 2025-11-26T15:41Z 4403 followers, 44.4K engagements

"You dont reach Staff Engineer by writing more code - you reach it by thinking at a higher altitude. Most engineers never make the jump because they keep improving their skills not their scope. To reach Staff-level early stop optimizing for speed and start optimizing for impact. Learn to see systems end-to-end debug across layers communicate clearly and make decisions that unblock entire teams not just yourself. Staff isnt about being the smartest coder in the room; its about being the engineer who reduces chaos increases clarity and constantly delivers leverage. --- Read the quoted post to"
X Link 2025-11-28T15:17Z 4401 followers, 318.4K engagements

"Some additional useful tips if you are a dev learning never stops in tech. focus on a little bit everyday so it compounds. consistency beats intensity. code for XX minutes daily non-negotiable. not tutorials. actual code. a small function a quick script a tiny fix. hands on keyboard. If practicing for interviews drill one pattern a day. sliding window monday. two pointers tuesday. recursion wednesday. small reps deep grooves. revisit fundamentals weekly. how does a hash map actually work what happens during a TCP handshake why is Big O what it is surface knowledge fades deep understanding"
X Link 2025-11-30T12:44Z 4402 followers, 156K engagements

"For someone following Karpathy since his early days teaching deep learning at Stanford I feel he is speaking something so cool here. Most "educational" content on YouTube and TikTok is entertainment wearing a learning costume. You watch you nod you feel productive. But nothing sticks. Nothing changes. Learning is supposed to feel like effort. Not a 10-minute explainer. Not a mass-produced "Learn React in XX days" video. You want the mental equivalent of sweating. Declare your intent upfront. Binary variable. Are you consuming to be entertained or to actually learn If you're honest you'll"
X Link 2025-11-30T12:50Z 4402 followers, 23.7K engagements

"I keep noticing this pattern in engineering teams. A developer hits a bug. Opens ChatGPT or Copilot. Pastes the error. Copies the fix. Moves on. No understanding. No curiosity. Just the green checkmark so the ticket can close. But then there are the engineers I genuinely admire. They pause. They read the stack trace. They ask why this broke in the first place. They dig into the source code instead of just the symptoms. They learn something about the system they'll carry forever. They fix the bug and three other things they noticed along the way. That's the difference between using AI as a"
X Link 2025-12-01T18:07Z 4399 followers, 285.5K engagements

"- started learning rust seriously this year - posted consistently online shared what I learned built an audience - was deep into trading before learned the hard way that discipline excitement - met some insanely smart and kind people who changed my thinking - understood a lot about myself strengths blind spots patterns - trying to take more ownership at work stepping into responsibility instead of waiting for it -pushing myself to become the kind of engineer who handles real production issues not just theory - embracing consistency over chaos -getting comfortable being uncomfortable this year"
X Link 2025-12-02T14:10Z 4402 followers, 14.1K engagements

"ML engineering that is essential for backend engineers model serving & inference APIs. you'll eventually serve a model. understand latency vs throughput tradeoffs. know when to use REST vs gRPC for predictions. batch inference vs real-time. cold start problems are real. feature stores & pipelines. models are useless without features. learn how feature stores work (Feast Tecton). understand online vs offline features. the gap between training data and serving data will bite you. vector databases & embeddings. search is becoming semantic. know how embeddings work conceptually."
X Link 2025-12-03T22:53Z 4402 followers, 25.3K engagements

"Everyone loves jumping into Design a system for 1M users. And most candidates immediately scream: "Kafka K8s Redis S3 CQRS Microservices" (You might wanna bookmark this as a developer who is actively interviewing rn.) Well this is exactly how you fail real-world engineering. The best engineers dont start with tech. They start with questions - the kind that reveal whether you actually understand scale correctness and business goals. --- Here are the questions strong engineers ask: X. What does 1M users mean Daily active Concurrent Peak traffic Reads vs writes distribution X. Whats the business"
X Link 2025-12-04T11:59Z 4401 followers, 100.7K engagements

"Techniques I'd master if I wanted to go from good backend dev dangerous distributed systems engineer staff-level impact. Don't Bookmark this. Rather learn it asap. X. CAP PACELC and Real-World Tradeoffs Not theory how each constraint shows up in production. X. Leader Election Algorithms Raft Paxos Zab and when consensus is not the answer. X. Failure Detection & Heartbeating Perfect vs eventually-perfect detectors phi-accrual. X. Idempotency & Exactly-Once Semantics The difference between delivery vs processing guarantees. X. Distributed Log Internals Kafka ISR Leader/Follower commit index"
X Link 2025-12-05T14:36Z 4403 followers, 46.6K engagements

"Interviewer at Meta: "How do we serve Llama-3 to 1000 concurrent users And why do we run out of memory (OOM) even if the model weights fit perfectly on the GPU" This is how you answer it: If the model fits but we still crash the culprit is the KV Cache. The GPU isn't just holding model weights; it has to store the Key and Value matrices for every token in the conversation history. X user = manageable but 1000 users long context windows = VRAM explosion. Standard memory allocation reserves contiguous blocks for the "max possible length" leading to massive fragmentation and wasted space. We can"
X Link 2025-12-06T13:25Z 4402 followers, 114.8K engagements

"Age of Databases (with categories) (R = Relational KV = KeyValue DOC = Document COL = Columnar GRAPH = Graph OLAP = Analytics DIST = Distributed SQL EMB = Embedded TS = Time-series) --- 🟥 Oracle DB (R) XX years (1979) 🟥 Teradata (OLAP) XX years (1979) 🟥 IBM Db2 (R) XX years (1983) 🟥 Sybase ASE (R) XX years (1988) 🟥 Microsoft SQL Server (R) XX years (1989) 🟥 PostgreSQL (R) XX years (1996) (roots 86) 🟥 MySQL (R) XX years (1995) 🟥 MariaDB (R) XX years (2009) 🟦 CockroachDB (DIST) XX years (2014) 🟦 TiDB (DIST) X years (2016) 🟩 Berkeley DB (KV/EMB) XX years (1991) 🟩 Redis (KV) XX years"
X Link 2025-12-08T08:04Z 4403 followers, 26.4K engagements

"I wired up a tiny actor runtime in Go pumped 10M messages through it and got 30M ops/s. Its not production-stable yet but the pattern feels right. In Go the actor model is simply: one goroutine owns some state and you only touch that state by sending it messages over a channel. No locks no shared mutable state no random goroutines poking into your structs. --- Core idea - Each actor = X goroutine + X mailbox (chan Msg) - Actor has private state inside the goroutine - Other parts of the program never access that state directly they just Send(msg) - The actor loops forever: for msg := range"
X Link 2025-12-08T09:27Z 4403 followers, 43K engagements

"Another pattern I really like in Rust thats simple but insanely powerful: Newtype + smart constructors instead of naked primitives Instead of passing String u64 etc. everywhere and hoping they follow business rules you wrap them in domain types that enforce invariants on creation. --- struct NonEmpty(String); impl NonEmpty fn new(s: String) - OptionSelf if s.trim().is_empty() None else Some(NonEmpty(s)) fn as_str(&self) - &str &self.0 --- Why this helps: You encode invariants in the type system any NonEmpty you see is guaranteed valid. Fewer if .is_empty() checks sprinkled everywhere."
X Link 2025-12-08T11:39Z 4398 followers, 2406 engagements

"S3 looks like a dumb bucket but its one of the most advanced distributed systems ever built. If you talk to the team ask them this: How do you design something that serves 150M RPS offers 11-nines durability avoids global bottlenecks and still feels boring + predictable to the entire internet The real questions Id love answered: How they prevent hotspots without coordination How they prove 11-nines durability How conditional writes scale globally What real-world customer bugs teach them Where object storage goes next (transactions AI-native metadata) The engineering behind S3s boring"
X Link 2025-12-08T14:02Z 4403 followers, 73.7K engagements

"lowkey think the problem wasnt DSA or web dev itself it was treating both like checklist grinding instead of using them to understand cs deeper you can still rebuild the foundation now: os + networks + db + distributed systems + a bit of math the upside now you also know how to ship things foundation built late is still a foundation"
X Link 2025-12-08T14:12Z 4403 followers, 10.8K engagements

"In engineering the same truth applies: The depth of your understanding is directly proportional to the amount of pain youre willing to endure. You dont become a great engineer by: - skimming docs - copy-pasting StackOverflow code or - watching system design in XX mins videos. You grow when you struggle: - When your distributed system keeps failing at X AM - When your Kafka consumer is stuck in an infinite rebalance loop - When your Rust lifetimes make you question reality - When your Go service deadlocks because you just added a goroutine - When your Kubernetes deployment breaks and nobody"
X Link 2025-12-09T08:14Z 4403 followers, 49.5K engagements

"@ChShersh - i8 stores X bits - IPv6 has X segments - UTF-8 uses variable bytes - JavaScript thinks + === "" At this point naming things is more of a collective prank passed down through generations 😂"
X Link 2025-12-09T16:08Z 4402 followers, XXX engagements

"🛠 Open Source Infra in Plain English (so you dont feel lost): MinIO S3 but open source (object storage) NATS super-fast messaging system (pub/sub + queues) Kafka event streaming backbone for big systems Etcd a tiny but mighty key-value store for coordination Consul service discovery yellow pages Redis in-memory cache that makes everything fast ClickHouse analytics database that flies PostgreSQL the reliable SQL database people actually love CockroachDB Postgres-compatible but globally distributed MariaDB/MySQL battle-tested relational databases Qdrant / Weaviate vector databases for AI"
X Link 2025-12-10T12:44Z 4403 followers, 7399 engagements