Private corporate batches delivered onsite across Bangalore, or live online in IST (UTC+5:30) — taught by a practitioner who runs Elasticsearch in production.
20 years across DevOps, SRE and Security · 10,000+ engineers trained · Trained teams at JPMorgan Chase, Verizon, Nokia and the World Bank
Principal DevOps Engineer & Architect
Rajesh teaches Elasticsearch at both ends: the analysis chain and mapping decisions that determine what a query can ever match, relevance work with BM25 explained through the explain API, function scoring and synonym management; and the operational half — shard sizing, allocation, hot node diagnosis, mapping explosions, index lifecycle management and data tiers. Sessions run against a real multi-node cluster where indices are deliberately mis-sized and mappings deliberately wrong, so attendees practise diagnosis and reindexing rather than only the happy path.
Twenty years across DevOps, SRE and Security, in principal and architect roles at PayPay, SoftwareAG, ServiceNow, JDA Software, Intuit, Adobe and others. He has trained engineers at JPMorgan Chase, Verizon, Nokia, the World Bank, VMware, Oracle, Mercedes-Benz and Airbus — more than 10,000 people personally. He teaches what he runs, not what he reads.
You are booked with a named engineer, and that is who turns up. Marketplaces and larger providers rotate whoever is free, so the person who sold you the agenda is rarely the person teaching it.
The same trainer is available for the next engagement, which matters when a team builds on what it learned last time.
Your batch is assigned a named trainer before it starts, and that is who teaches it. See the full faculty.
Engagements are matched on the tool, not the calendar. For Elasticsearch that means a trainer who has run it in production — Elasticsearch for Bangalore search and observability teams — analysis chains, relevance tuning, shard sizing and index lifecycle cost control — rather than whoever is free that week. You are told who is teaching before you commit, and that person is on the discovery call that shapes the agenda.
Where a batch is large enough to need a second trainer, the pairing is declared up front. The lead trainer stays accountable for the syllabus and the assessment either way.
Principal DevOps Engineer & Architect
Twenty years across DevOps, SRE and Security in principal and architect roles at PayPay, SoftwareAG, ServiceNow, JDA Software, Intuit, Adobe, IBM/Emptoris, Ness, MindTree and Accenture. He has trained more than 10,000 engineers personally, at organisations including JPMorgan Chase, Verizon, Nokia, the World Bank, VMware, Oracle, Mercedes-Benz and Airbus. He teaches what he runs, not what he reads.
Teams of 8–30
Custom agenda, your timezone, onsite or online, NDA-friendly.
Request a quoteIndividual engineers
A private instructor and a curriculum built around your goal.
₹99,999Individuals who want peers
Scheduled batch, max 8 to 10 hours of live instruction.
₹34,999A private batch starts with a discovery call. We look at the stack you actually run — the CI system, the cloud, the constraints — and map the agenda onto it, so examples use your topology rather than a generic one.
Onsite sessions run at your own office. Elasticsearch engagements in Bangalore commonly land in Koramangala and HSR Layout for product search teams, and along the Outer Ring Road, in Whitefield or at Manyata Tech Park for platform and observability teams; we schedule one site per day so a cross-city move does not consume lab time. You provide the room, a screen and network access; attendees run a multi-node cluster locally in containers or on their own cloud accounts, and where the engagement is about your relevance or your shard layout we work against an anonymised sample of your index under NDA rather than against production. Sessions are scheduled in IST (UTC+5:30) and planned around release and peak-traffic windows — search teams here usually avoid sale periods entirely. Invoicing is in INR from our Indian entity with GST, against a purchase order where procurement requires one, and travel within Bangalore is included in the quote.
Every attendee leaves with recordings, slides, lab repositories and a completion certificate. You receive an attendance and assessment report. Invoicing supports PO and GST.
Talk to us about a private Elasticsearch batchElasticsearch is a distributed document store whose defining feature is what happens to text on the way in. Every field passes through an analysis chain — character filters, a tokeniser, then token filters — and what that chain emits is what the inverted index actually contains. A query is analysed the same way before it is matched, so search behaviour is decided by two chains agreeing. Almost every relevance complaint traces back to an analyser that was chosen by default rather than designed, and almost every relevance improvement is a change to tokenisation, normalisation, synonyms or stemming rather than to the query.
Mapping is the second decision that is hard to undo. A field's type, whether it is indexed, whether doc values exist for sorting and aggregation, and whether a keyword sub-field is available for exact matching are all fixed when the index is created; changing them means reindexing. Dynamic mapping makes it easy to create a field per unique key in a log payload, which is how clusters end up with tens of thousands of fields and a cluster state that is expensive to distribute.
Operationally, an index is a set of shards, each an independent Lucene index, allocated across nodes and replicated for availability. Shard count is set at creation, shard size drives heap pressure and recovery time, and allocation decisions are the source of most cluster incidents. On top sit the two query modes — scoring query context using BM25, and cacheable yes-or-no filter context — plus the aggregation framework that turns the same data into analytics. Index lifecycle management, rollover and data tiers are what keep all of that affordable as the data keeps arriving.
Search and log data have both outgrown the naive setup that got teams started. A product search index that worked at fifty thousand items behaves differently at ten million, and a logging cluster sized for one team becomes a shared platform whose cost is visible to finance. In both cases the failure is the same: nobody made a deliberate decision about analysis, mapping, shard count or retention, and by the time the symptoms appear the fixes require reindexing.
Relevance has also become a measurable business concern rather than an engineering preference. Teams now run offline evaluation against judgement sets, watch conversion by query segment, and treat a synonym change as a release. That requires people who understand scoring — why BM25 ranked a document where it did, what function scoring does to that, when to use a rescore window — rather than people who can call a match query.
On the operations side, the licence change that produced the OpenSearch fork left every team with an architecture decision they cannot dodge, especially teams building in the Mumbai region where a managed OpenSearch service sits alongside Elastic Cloud. Choosing between them, and knowing which features and APIs are actually shared, is now part of the job. Between relevance engineering, cluster economics and that platform choice, the demand has shifted decisively from familiarity towards judgement.
The model underneath everything. Documents, indices, shards and replicas; nodes and their roles; how a write is routed, indexed and made searchable; and the refresh, flush and translog behaviour that explains why a document you just wrote is not there yet.
Topics: Documents, indices, shards and replicas · Node roles: master, data, ingest, coordinating · Routing a write and the primary-replica path · Refresh interval, translog and flush · Near real time: what near actually means · Reading cluster health and node stats
Text analysis as design work. Character filters, tokenisers and token filters; standard, keyword, language and custom analysers; the analyze API for proving what a chain emits; and the specific problems of mixed-script, transliterated and abbreviation-heavy catalogues.
Topics: Character filters, tokenisers and token filters in sequence · Built-in analysers and their real behaviour · Custom analysers and per-field analysis · Search-time versus index-time analysis and when they must differ · Synonyms, stemming, stop words and normalisation · Phonetic, transliteration and ASCII folding for mixed-language data · The analyze API as a debugging tool
What you cannot change later. Explicit versus dynamic mapping, core field types, text versus keyword and multi-fields, doc values, nested versus object, index templates and component templates — then aliases and the reindex API for changing what is supposedly fixed.
Topics: Explicit mapping, dynamic mapping and dynamic templates · text, keyword, numeric, date, boolean and multi-fields · Nested versus object and the queries each enables · doc_values, index and store options · Index templates, component templates and data streams · Mapping explosions and field count limits · Aliases and zero-downtime reindexing
Asking questions precisely. Query context versus filter context; term-level versus full-text queries; bool composition with must, should, filter and must_not; phrase, prefix and fuzzy matching; and the debugging tools that show what actually ran.
Topics: Query context, filter context and the request cache · term, terms, range, exists and prefix queries · match, match_phrase, multi_match and query_string · bool composition and how clauses combine · Fuzziness, edit distance and where it fails on Indian-language data · Pagination: from and size, search_after and point in time · profile and explain for diagnosing behaviour
Making results better on purpose and proving it. BM25 and its parameters, field boosting, function score with business signals, rescoring, synonym governance, and an evaluation loop using judgement sets so a relevance change can be defended with numbers.
Topics: BM25 scoring and reading an explain output · Field boosting and multi_match types · function_score, decay functions and business signals · Rescore windows and two-phase ranking · Synonym files versus synonym graph, and managing them over time · Search relevance evaluation with the ranking evaluation API · Handling misspellings, abbreviations and mixed-script queries
The same index answering analytical questions. Metric and bucket aggregations, nesting them, date histograms and time series, cardinality and its accuracy trade-off, pipeline aggregations, and the memory and performance consequences of asking for too much.
Topics: Metric aggregations and their exactness · Bucket aggregations: terms, range, date_histogram, filters · Nesting aggregations and sub-aggregation ordering · cardinality, percentiles and approximation trade-offs · Pipeline aggregations: derivative, moving function, bucket_script · Aggregations on nested documents · Memory pressure, circuit breakers and expensive aggregations
Running it in production. Shard sizing rules that hold up, allocation and rebalancing, hot node diagnosis, JVM heap and garbage collection, snapshot and restore, rolling upgrades, and the small set of APIs that answer almost every incident question.
Topics: Shard sizing, count and the cost of too many · Allocation, awareness, filtering and explain API · Hot nodes, thread pools and rejected requests · Heap sizing, field data and circuit breakers · Snapshot repositories, restore and testing a restore · Rolling upgrades and version compatibility · Cluster health triage: red, yellow and what to do first
Getting data in and keeping the bill sane. Ingest pipelines and processors, Beats and Logstash, data streams and rollover, index lifecycle management across hot, warm, cold and frozen tiers, retention policy, and an honest comparison of self-managed Elasticsearch, Elastic Cloud and OpenSearch.
Topics: Ingest pipelines, processors and simulate · Beats, Logstash and when each is appropriate · Data streams, rollover and time-series indices · Index lifecycle management across data tiers · Searchable snapshots and frozen tier economics · Retention policy and what to tell the team that wants everything forever · Self-managed, Elastic Cloud and OpenSearch: licence, feature and API differences · Security basics: authentication, roles and field-level access
We rebuild the agenda around the tools you actually run.
Request a custom agendaBuild a custom analysis chain for a catalogue containing transliterated brand names, abbreviations and mixed English and Indic script, and prove the tokens with the analyze API.
Reproduce a cluster state blown up by dynamic fields from a log payload, contain it with dynamic templates, then migrate the index behind an alias with no downtime.
Take ten real queries with poor results, improve them with boosting, function scoring and synonyms, and defend the change with a ranking evaluation run rather than an opinion.
Profile a slow search, move non-scoring clauses into filter context, fix pagination with search_after, and record the latency and cache-hit difference.
Diagnose an unbalanced cluster with a hot node and unassigned shards using the allocation explain and cat APIs, correct it, then snapshot and restore to verify recovery.
Design ingest pipelines, data streams and a lifecycle policy across hot, warm and cold tiers for a real log volume, then measure the storage and query impact of the change.
Elasticsearch in Bangalore splits cleanly into two jobs, and the teams doing them rarely talk to each other. The first is product search. A large share of India's consumer commerce, food delivery, travel and classifieds engineering sits in Koramangala, HSR Layout, Bellandur and Whitefield, and for those teams relevance is a revenue metric — which means the interesting work is in the analysis chain rather than the cluster. Indian catalogues arrive with brand names transliterated three ways, mixed English and Indic script in the same query, aggressive abbreviation, and misspellings that a plain edit-distance fuzzy match handles badly. Custom analysers, synonym management, phonetic and transliteration filters, function scoring against business signals and a way to evaluate whether a relevance change actually helped are the real syllabus for that audience.
The second job is observability and investigation. The GCC engineering sites along the Outer Ring Road and at Manyata Tech Park run Elasticsearch as the log and event store behind a shared platform, and the problems there are cluster problems: shard counts that grew without a plan, hot nodes, mapping explosions caused by dynamic fields in someone's log line, and a storage bill that grows faster than the estate. Index lifecycle management, data tiers, rollover, field mapping discipline and retention policy dominate those engagements. A third question now sits on top of both: teams building on AWS in the Mumbai region regularly have to choose between self-managed Elasticsearch, Elastic Cloud and the OpenSearch service, and the licence and feature differences are a design decision rather than a preference. Bangalore job specs for this skill increasingly ask for one of these two profiles specifically, not for generic familiarity.
Billed yearly at ₹9,996
Max 8 to 10 hours of live instruction
Full program, private instructor
8–30 engineers · custom agenda · onsite or online · PO and GST invoicing
Refunds. If we cancel or postpone a cohort, you get a full refund within 15 days. There is no money-back guarantee otherwise.
Terms. Course material remains licensed to the attendee. Read the terms.
Your data. We don't share it with third parties. Privacy policy.
Certificate of completion
I took Terraform training with the tutor named Mithilesh. I requested to tailor the course curriculum for my needs. He did an excellent job of showing me how to write the Terraform script per the instructions provided.
My experience with the AIOps training was positive. The course covered important topics in a structured way, and Rajesh Kumar explained the concepts patiently. I found the practical aspects particularly helpful because they made the technical content easier to understand.
I was looking to improve my understanding of AIOps, and this training helped me achieve that goal. Rajesh Kumar explained the subject in a structured and practical manner. The sessions on different AIOps concepts were informative.
I recently did a SRE Session with Rajesh Kumar from DevOps School and the session was great. Right from 1st day till day 15, we had a very interactive session. Rajesh clarified our doubts and the tool demos were excellent without any hiccups. He simplified the concepts while sticking to the content with a fine balance between theory and practice. Am convinced he is one of the best trainers for SRE & DevOps concepts.
The Rundeck developer session was excellent and highly engaging. I appreciated how well the session was structured, with the theoretical concepts explained clearly and in simple terms. What stood out most to me was the demo — it was both informative and enjoyable. I especially liked how Rajesh walked us through not only the happy path but also the sad path, showcasing common issues and sharing practical troubleshooting tips.
Rajesh's experience and knowledge are exceptional and we learnt invaluable practical knowledge which we can apply in our production environment. Incredibly friendly and gave us a fantastic insight both in-depth and at a high level of the Rundeck product.
| What matters | YouTube + blogs | Generic online course | Freelance marketplace | DevOpsSchool |
|---|---|---|---|---|
| Named practitioner | No | Rarely | Varies per booking | Yes — same trainer each time |
| Production experience | Unknown | Unknown | Unverified | 20 years, named employers |
| Custom agenda | No | No | Sometimes | Built from your stack |
| Onsite delivery | No | No | Sometimes | Yes |
| Lab environment | None | Sandbox that expires | Varies | Your own cloud — skill goes with you |
| Assessment | None | Quiz | Rarely | Assignments + capstone per module |
| Per-attendee certificates | No | Sometimes | Rarely | Yes |
| Corporate invoicing | No | Limited | Varies | PO and GST |
| Post-training support | None | Forum, time-limited | None | Lifetime forum access |
Tell us the team, the stack and the timeline. You'll get a straight answer, not a sales sequence.
Talk to an advisorPrefer to call or email?
More ways to reach us on the contact page.