<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Ashutosh Tiwari on Medium]]></title>
        <description><![CDATA[Stories by Ashutosh Tiwari on Medium]]></description>
        <link>https://medium.com/@ashutosh122002?source=rss-9b26e540bde0------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*ky28_IjwMum_F_Wlh-NKaw.jpeg</url>
            <title>Stories by Ashutosh Tiwari on Medium</title>
            <link>https://medium.com/@ashutosh122002?source=rss-9b26e540bde0------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 01:42:05 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@ashutosh122002/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Tokenization Supercycle- Private markets could be next]]></title>
            <link>https://medium.com/@ashutosh122002/tokenization-supercycle-private-markets-could-be-next-4ca5f5183d81?source=rss-9b26e540bde0------2</link>
            <guid isPermaLink="false">https://medium.com/p/4ca5f5183d81</guid>
            <category><![CDATA[rwa]]></category>
            <category><![CDATA[tokenization]]></category>
            <category><![CDATA[private-markets]]></category>
            <category><![CDATA[solana-network]]></category>
            <category><![CDATA[blockchain]]></category>
            <dc:creator><![CDATA[Ashutosh Tiwari]]></dc:creator>
            <pubDate>Thu, 01 Oct 2026 02:26:38 GMT</pubDate>
            <atom:updated>2026-10-01T02:28:59.799Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*U83Ilxiq6jUHj18H4ChqEg.png" /></figure><p>Private are getting bigger more valuable, the problem is even though they are private owning a piece is still difficult.</p><p>According to S&amp;P Global private market asset under management (AMU) grew from around 10.89 trillion in 2022 to roughly 15 trillion in 2024, and are progressing to exceed 18 trillion by 2027.</p><p>This is a clean and clear metric of how fast private markets are growing.</p><p>In a research by chief Economist of Apollo Tristen Slok found that 87.8 % of US firms with having revenue more than 100 million dollars are private.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*UUQzxDQtiAi_XSAR.jpg" /><figcaption>Source: <a href="https://www.apolloacademy.com/many-more-private-firms-in-the-us/">https://www.apolloacademy.com/many-more-private-firms-in-the-us/</a></figcaption></figure><p>Getting a single number for the entire private equity market is difficult because there isn’t a clean global figure that captures everything.</p><p>A better pov is venture backed unicorns. PitchBook counted around 1,700 private companies valued at $1 billion or more, together worth about $8.6 trillion in March 2026. For comparison, that group had only<strong> </strong>228 companies with a combined value of roughly $800 billion in 2016.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/898/1*RsXSaMsu_2n6gYljVKoI5A.png" /><figcaption>Source: PitchBook data as of March 31, 2026</figcaption></figure><p>Private ownership has became enormous. The growing share of valueable companies are staying private. Before, taking about tokenziation lets have a look at what happens when ownership stays private at this scale.</p><p><strong>Who can currently buy private company shares ?</strong></p><p>The problem is not that these companies are private, the problem is that a large part of market is restricted to a specific class of investors.</p><p>In US people with more than $200k or $300k with a spouse or people with more than $1 million net worth excluding their primary residence and PE funds can only participate in buying. Also, according to a research by sacra.com, accredited investors definition in 2024 is increased to 13%.</p><p>This clearly shows that for someone who doesn’t meet these requirements, direct access to private company shares remains limited. Only funds provide the main route to participate.</p><p>Now, the question is can existing share holders easily sell?</p><p>If a share holder wants to transfer the shares to a buyer, its a formal process that includes a verification and approval of the company itself and the company has a right of first refusal(ROFR), it can buy the shares at the agreed price from the share holder itself. So, it is not easy to transfer.</p><p>If the share has to happen, the company itself approves that. It initially shares the share holder agreement, and most have restrictions on the share transfer. Just like the share holder, the buyer should also need to be accredited and should fulfill the desired conditions as well.</p><p>Slowly, a secondary trade takes days to months, legal documents are drawn up, funds move, and the cap table is finally updated. Also, friction multiplies when the buyer and seller are in different countries.</p><p>If I conclude, these points are the main difficulties that needs to be improved -<br>Access, consent, records, settlement speed, location constrained.</p><p><strong>Introduction to Tokenization -</strong></p><p>Tokenization turns ownership of assets like bonds, funds, private companies, and real estate into blockchain-based tokens. The asset doesn’t change the way it is recorded, transferred, and settled does.</p><p>RWA market size refers to the total value of such tokenized assets recorded onchain at a given point in time.</p><p>The sector gained major traction after BlackRock launched its BUIDL fund through Securitize on Ethereum in March 2024. It grew from $40 million at launch to more than $1.8 billion onchain by late 2025.</p><p>Then Robinhood brought tokenized U.S. stocks and ETFs to European users on Arbitrum, pushing tokenized assets beyond institutions and closer to retail investors.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*cXh-wpMVKa5Nt3ugiAN9mQ.png" /></figure><p>Private markets rely entirely on separate systems for issuing securities, maintaining ownership records, transferring shares, and settling transactions, some of these processes can move into tokens.</p><p>On May 2026, the first fully regulated issuer sponsored tokenized equity stack was launched on Solana.</p><p>Solana&#39;s top DEXs securitize, jump trading, and Jupiter created one transaction path, by combining three separate layers of transaction.</p><p><strong>Securitize</strong> -&gt; broker-dealer/ATS and transfer-agent infrastructure. <br><strong>Jump </strong>-&gt; PropAMM liquidity engine. <br><strong>Jupiter&#39;s </strong>-&gt; Solana-wide order routing.</p><p>All three combines to help a trade in a tokenized equity in the same 400 millisecond inside a block in Solana.</p><p>In an official statement, it is stated that tokenization is divided into two types -<br>1. issuer-sponsored tokenization - (the issuer mints the token)<br>2. third party sponsored tokenization- (a third party approves it)</p><p>Securitize issued shares fall in the first.</p><p>The pre-launched tokenized security has zero liquidity, but the bid asks spreads 10-50 points, more than 10x what a costumer pays on a public stock.</p><p>According to a report by fxempire before any public listing Anthropic&#39;s pre IPO valuation was $1 Trillion through onchain activity on Jupiter.</p><p>Coming to the above problems, who approves the transfer and how are records maintained?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*iUntg6YP4foCYXAWjp6hag.avif" /><figcaption>Source: RWA.xyz</figcaption></figure><p>Private share ownership also comes with the challenge of maintaining ownership records and processing transfers within company and regulatory rules.</p><p>This is the most interesting part about tokenization infrastructure.</p><p>In March 2026, NYSE partnered with Securitize on its upcoming Digital Trading Platform, naming Securitize its first digital transfer agent for blockchain securities. A month later securitize partnered with Computershare to let participating US listed companies issue tokenized shares representing direct equity ownership.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*JBAuKimAYrEuUN4MTDS7hA.avif" /><figcaption>source: RWA.xyz</figcaption></figure><p>These developments show that tokenization isn&#39;t just about putting shares on a blockchain. It also involves rebuilding how securities are issued, recorded, and transferred.</p><p>Distribution of Tokenized RWA reached from $ 11.8 billion to $38.8 billion as of sept 2026. That’s a 228.8 % increase in roughly in a year with holders above 1.5 million as of August 2026.</p><p>What’s still unsolved is Tokenization doesn&#39;t automatically make private shares accessible to everyone. Investors still need to pass KYC and meet the required eligibility criteria, such as the accredited investor list.</p><p>In other words, Tokenization can change how ownership is handled, but it doesn&#39;t by itself change who is legally allowed to own it.</p><p>So if tokenization doesn’t solve the entire private market problem by itself, what would actually create a tokenization supercycle?</p><p><strong>1. Regulatory clarity</strong>-<br>Who is allowed to buy them? Can they be traded 24/7? How is ownership legally recognized? What happens across countries?</p><p><strong>2. Secondary market liquidity</strong><br>A secondary market is where existing holders can sell to other investors. if more buyers find seller and trade is easily made then liquidity increases.</p><p><strong>3. Diversification beyond tokenized Treasuries</strong><br>The thesis can be much bigger if tokenization can expand to private-company shares, real estate funds, commodities credit.</p><p>With a rapid speed its growing, A common prediction in the RWA industry is that tokenized assets could reach a trillion dollar market by 2030.</p><p>Resources: <br><a href="https://thedefiant.io/news/defi/rwas-became-wall-street-s-gateway-to-crypto-in-2025">https://thedefiant.io/news/defi/rwas-became-wall-street-s-gateway-to-crypto-in-2025</a></p><p><a href="https://www.stobox.io/blog/rwa-market-size-2030-trillion-projections">https://www.stobox.io/blog/rwa-market-size-2030-trillion-projections</a></p><p><a href="https://blog.portals.fi/tokenized-rwa-growth-2026/">https://blog.portals.fi/tokenized-rwa-growth-2026/</a></p><p><a href="https://thedefiant.io/news/defi/tokenized-real-world-assets-cross-usd20-billion-in-onchain-value">https://thedefiant.io/news/defi/tokenized-real-world-assets-cross-usd20-billion-in-onchain-value</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4ca5f5183d81" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to avoid losing money to MEV bots -]]></title>
            <link>https://medium.com/@ashutosh122002/how-to-avoid-losing-money-to-mev-bots-2810c6c9ab32?source=rss-9b26e540bde0------2</link>
            <guid isPermaLink="false">https://medium.com/p/2810c6c9ab32</guid>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[transactions]]></category>
            <category><![CDATA[mevbot]]></category>
            <category><![CDATA[mev]]></category>
            <category><![CDATA[crytocurrency]]></category>
            <dc:creator><![CDATA[Ashutosh Tiwari]]></dc:creator>
            <pubDate>Sun, 27 Sep 2026 18:39:24 GMT</pubDate>
            <atom:updated>2026-09-27T18:42:33.168Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Tw_fBfTH8nRbKJ9228Njjw.png" /></figure><p>Before diving straight into the problem I am assuming that you are well<br>aware of slippage and MEV attacks and traded in crypto before.</p><p>Why this happens is a good thing to first understand, without going into<br>any definitions and explainer I would say it straight.</p><p>This happens generally in low liquidity pools where, shortage in supply results in increase in the assets price.</p><p>The main cause of slippage and sandwich attacks is, someone else(generally bots) putting their trade before yours when your trade is waiting on the mempool before getting executed.</p><p>The trick is, if someone is buying ahead of you this increases the token prices which results in a receiving little lesser tokens and then immediately sending them to pocket the difference in the increased price. These are called sandwich attacks.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WnG1hSNWCJj3rFKP_1Q65Q.png" /></figure><p>I’ve seen the usual advice that is to increase slippage when a transaction keeps reverting. That’s useful for getting a transaction through. But you’re also increasing the range in which your trade can execute.</p><p>For e.g. If your swap is expected to return $10,000:</p><p>0.5% slippage → minimum → $9,950<br>2% slippage → minimum → $9,800</p><p>That doesn’t mean a bot will automatically sandwich you at 2%. It means you’ve given the transaction more room to execute at a worse price. I see slippage as an execution setting, not MEV protection.</p><p>A $10,000 swap isn’t equally risky across every pool. If the pool has deep liquidity, my trade might barely move the price.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yfw8rWDGk3b3dDzsMG1m6g.png" /></figure><p>If the pool is shallow, the same $10,000 can move the pool significantly. And the larger the price movement created by my own trade, the more interesting that trade can become to an MEV searcher.</p><p>Suppose, we take two routes which are -</p><p>Route A<br> ETH → USDC<br> Expected: $9, 950<br> Price impact: 0.2%</p><p>Route B<br> ETH → TOKEN → USDC<br> Expected: $10, 020<br> Price impact: 1.8%</p><p>Route B here is giving a better quote, but you are taking a much larger price impact here. This shows that the highest quoted output isn’t always the safest execution.</p><p>I would also not suggest to use the same slippage tolerance for a liquid ETH/USDC pool and a thin altcoin pool. The first can absorb my trade without moving much. The second can’t.</p><p>better thing to do is to use - <br>liquidity + trade size + volatility + slippage together instead of treating slippage as a standalone number.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-SvN96KqYb4reGsBZNiVdA.png" /><figcaption>this is a sample image just for clarification</figcaption></figure><h4>How to avoid losing money to MEV bots -</h4><p>1. Not to allow pending transaction with visibility -<br>have a private/protected execution</p><p>2. Don’t give the trade unnecessary execution range.<br>More Slippage</p><p>3. Don’t trade large amount into thin liquidity.</p><p>Things to check before clicking swap -</p><p>Slippage<br>Price impact<br>Pool liquidity<br>Route<br>Transaction protection<br>Trade size relative to liquidity</p><p>Understanding these six things properly would help you avoid many of the unecessary losses that happen during execution.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2810c6c9ab32" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Building a Wallet Tracker with Alchemy Data API -]]></title>
            <link>https://medium.com/@ashutosh122002/building-a-wallet-tracker-with-alchemy-data-api-a0d7f85c5190?source=rss-9b26e540bde0------2</link>
            <guid isPermaLink="false">https://medium.com/p/a0d7f85c5190</guid>
            <category><![CDATA[indexing]]></category>
            <category><![CDATA[api]]></category>
            <category><![CDATA[rpc]]></category>
            <category><![CDATA[alchemy]]></category>
            <dc:creator><![CDATA[Ashutosh Tiwari]]></dc:creator>
            <pubDate>Sat, 19 Sep 2026 11:54:29 GMT</pubDate>
            <atom:updated>2026-09-19T11:54:29.934Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*mGTkdYXlQNVDxbS2ENSclQ.png" /></figure><p>In this day and age of AI. I built a simple wallet tracker from scratch, just for testing my skills. As, since the past few years I am only vibe coding. So, it was difficult in the beginning, but eventually I figured it out.</p><p>I was trying to build an app which on opening and entering the wallet address, returns -<br>1. How much ETH it owns<br>2. which tokens it holds<br>3. Recent transfers<br>4. NFTs<br>5. Transaction history</p><p>It looked simple, but when I started building it. I got to know that fetching all the data, the wallet holds for the given address isn’t an easy task. You can easily get the current state of a wallet but not the historical data.</p><p>If a user enters his wallet address in that application, it returns with the current state of tokens the address has. Simply, the RPC request used to fetch the balance is eth_getBlanace(walletAddress).</p><p>But only getting the balance doesn’t finishes the job, application needs to show more than that.</p><p>Ethereum does not have a simple RPC call for requesting the total tokens a specific wallet address owns. It has the record of all the transactions and state changes, and I need to create a pipeline that carries that data and shows in my application.</p><p>Info like what tokens/NFTs does this wallet own right now, can be fetched for creating a pipeline that connects to the blockchain history for a specific address<br>Like -</p><p>Wallet received 500 USDC.<br>Wallet sent 200 USDC. <br>Wallet received 1 ETH.<br>Wallet interacted with Contract X.</p><p>One way to achieve this, is to build an indexer. But, using an indexer takes a lot time to maintain block ingestion event processing, historical storage, indexing pipelines, token metadata etc. I needed a way to query already-indexed blockchain data.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*H6yDu8JDoxcxHtUyPPoLGg.png" /></figure><p>This lead me to found about Alchemy’s Data APIs.</p><p>These APIs are pre-indexed blockchain data that can be used for wallets, portfolios apps, dashboards and other applications. Instead of indexing everything.</p><p>Now, the application can access the data directly without indexing.</p><p>For example, I could send a request like this -</p><pre>const response = await fetch(<br>  `https://eth-mainnet.g.alchemy.com/v2/${ALCHEMY_API_KEY}`,<br>  {<br>    method: &quot;POST&quot;,<br>    headers: {<br>      &quot;Content-Type&quot;: &quot;application/json&quot;<br>    },<br>    body: JSON.stringify({<br>      jsonrpc: &quot;2.0&quot;,<br>      id: 1,<br>      method: &quot;alchemy_getAssetTransfers&quot;,<br>      params: [<br>        {<br>          fromBlock: &quot;0x0&quot;,<br>          toAddress: walletAddress,<br>          category: [<br>            &quot;external&quot;,<br>            &quot;erc20&quot;,<br>            &quot;erc721&quot;,<br>            &quot;erc1155&quot;<br>          ],<br>          withMetadata: true,<br>          excludeZeroValue: true<br>        }<br>      ]<br>    })<br>  }<br>);<br><br>const data = await response.json();<br><br>console.log(data.result.transfers);</pre><p>And the response is already structured around the information my application needs:</p><pre>{<br>  &quot;transfers&quot;: [<br>    {<br>      &quot;from&quot;: &quot;0x123...&quot;,<br>      &quot;to&quot;: &quot;0x456...&quot;,<br>      &quot;value&quot;: 500,<br>      &quot;asset&quot;: &quot;USDC&quot;,<br>      &quot;category&quot;: &quot;erc20&quot;,<br>      &quot;hash&quot;: &quot;0xabc...&quot;,<br>      &quot;blockNum&quot;: &quot;0x123456&quot;<br>    },<br>    {<br>      &quot;from&quot;: &quot;0x456...&quot;,<br>      &quot;to&quot;: &quot;0x789...&quot;,<br>      &quot;value&quot;: 1,<br>      &quot;asset&quot;: &quot;ETH&quot;,<br>      &quot;category&quot;: &quot;external&quot;,<br>      &quot;hash&quot;: &quot;0xdef...&quot;,<br>      &quot;blockNum&quot;: &quot;0x123457&quot;<br>    }<br>  ]<br>}</pre><p>I didn’t build the system that found these transfers.</p><p>I just queried the indexed data.</p><p>I needed my application to show three major things -&gt;<br>What tokens and their balance does my wallet holds?<br>Previous transaction history?<br>Does the NFT holds an NFT?</p><p>For token balances, I used Alchemy’s toke APIs that can return ERC-20 balances for an address. Its newer Portfolio/Data APIs can also return token balances with metadata and prices across supported networks.</p><p>The application can now show this directly -</p><p>Let’s have threee parameters here — tokens, value, balance.</p><p>ETH’s value is 1.42 for a total $6,000<br>USDCs value is 520 for a total of $520<br>LINK’s value is 32 for $500</p><p>For activity feed and transaction history. I used Alchemy’s Transfer API , which can retrieve the historical asset transfers for any address.</p><p>The application can choose, which information it needs from the blockchain and run that specific Alchemy API in accordance to that request.</p><p>Without having a system that manages the indexed data layer itself, I needed to build the whole structure myself, and that is simply building a new architecture inside the architecture, I am trying to build.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*C3x-ubUGhJeiWnQUH6bCZg.png" /></figure><p>And RPC is still needed, for submitting a transaction reading a specific piece of chain state, interacting with a smart contract. RPC and indexed data solve different problems.</p><p>RPC is the access to the chain, indexing system is to fetch the data directly from the chain.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a0d7f85c5190" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[I Tested 7 free public RPC’s, here’s what I found -]]></title>
            <link>https://medium.com/@ashutosh122002/i-tested-7-free-public-rpcs-here-s-what-i-found-162192308b62?source=rss-9b26e540bde0------2</link>
            <guid isPermaLink="false">https://medium.com/p/162192308b62</guid>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[rpc-node]]></category>
            <category><![CDATA[infra]]></category>
            <category><![CDATA[rpc]]></category>
            <dc:creator><![CDATA[Ashutosh Tiwari]]></dc:creator>
            <pubDate>Fri, 11 Sep 2026 18:30:46 GMT</pubDate>
            <atom:updated>2026-09-11T18:41:47.462Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*brKKAxRJd8ZQtPHBPNWVMw.png" /></figure><p>You are building an app and your budget is zero and you need a reliable node connection.</p><p>This is the exact phase most of the devs go through. And I am one of them, its called “hunting for free public RPCs” haha just kidding 😂.</p><p>But nothing here is free, everything comes with a cost one or the other way around.</p><p>In my previous articles, we have discussed about the limitations of RPC providers. Out of these the fastest one is rate limiting.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7ru1Ep4yKESrHToq0Od39Q.png" /></figure><p>Using free infrastructure felt good, until the app starts dropping transactions, returning rate limit errors or taking more time to fetch a block header.</p><p>I got tired of using this hit and trial with each RPC provider in the search of the perfect one. SO, I set up a script and listed these 7 free RPC endpoints.</p><p>1. dRPC<br>2. PublicNode<br>3. Ankr<br>4. Lava Network<br>5. Nodies<br>6. Chainstack(Public tier)<br>7. 1RPC</p><p>I actually tested 7 RPC endpoints — <br>1. Rate Limiting<br>2. Privacy and security<br>3. Lowest Latency</p><p>1. Rate Limiting -<br>The most common error as soon as you start sending more requests, it returns with “HTTP-429 too many requests”. This is where free RPC infrastructure providers can start to fail.</p><p>My pick after trying above providers would be -<br>&gt; Ankr and Chainstack<br>&gt; Lava and Nodies</p><p>2. Privacy and security<br>If your users are sending requests directly through an RPC provider, the provider may be able to see their IP address along with the requests they are making. You also expect privacy from a provider.</p><p>for privacy, 1RPC stood out amongst the above.</p><p>3. Lowest Latency -<br>This is also, a big issue. If the users are using an app which is responding after few seconds, that wouldn’t be impressed. Even millisecond delay can spoil the whole ux.</p><p>For lowest latency I would choose dRPC and PublicNode. Returning the response withing under 30 milliseconds on standard calls from my test environment. Clean, fast and are great for basic requests.</p><p>Best Overall for Apps &amp; Frontends -<br>1. dRPC ([<a href="https://eth.drpc.org](https://eth.drpc.org)).">https://eth.drpc.org](https://eth.drpc.org)).</a><br>It had the best latency in my testing and handled my test traffic well.</p><p>Best Backup / Fallback Endpoint: <br>2. PublicNode ([<a href="https://ethereum.publicnode.com/">https://ethereum.publicnode.com</a>] (<a href="https://ethereum.publicnode.com/">https://ethereum.publicnode.com</a>)). <br>It performed consistently in my tests, making it a useful fallback when the primary endpoint is unavailable.</p><p>Best for Privacy: <br>3. 1RPC ([<a href="https://1rpc.io/eth](https://1rpc.io/eth)).">https://1rpc.io/eth](https://1rpc.io/eth)).</a> <br>Use this if you don’t want your users RPC requests getting tracked.</p><p>These are the results from my findings. Let me know, how it worked out for you.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=162192308b62" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[My App keeps showing Transaction Pending -]]></title>
            <link>https://medium.com/@ashutosh122002/my-app-keeps-showing-transaction-pending-04038958bf8a?source=rss-9b26e540bde0------2</link>
            <guid isPermaLink="false">https://medium.com/p/04038958bf8a</guid>
            <category><![CDATA[transactions]]></category>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[rpc]]></category>
            <category><![CDATA[html]]></category>
            <category><![CDATA[polling]]></category>
            <dc:creator><![CDATA[Ashutosh Tiwari]]></dc:creator>
            <pubDate>Thu, 10 Sep 2026 16:14:45 GMT</pubDate>
            <atom:updated>2026-09-10T16:31:45.089Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*o9ae-6M6wLf8tAZ2YGKfmg.png" /></figure><p>I built this app to check on the status of my transactions and got to know the real science behind how my it keeps on responding with transaction’s pending, even after I received the transaction has gone through onchain.</p><p>I have seen this in other apps too. Even after my tokens have been sent the sending transaction message keeps on looping on the screen.</p><p>Let’s see how-</p><p>My app is interacting with the blockchain here. When I click on send transaction an RPC request is sent to the blockchain node.</p><p>The blockchain node receives the request, processes it and sends back the transaction hash to the app. The transaction hash is not the confirmation of the transaction it is just an identifier that your transaction is submitted to the blockchain.</p><p>Confirmation happens later when the transaction is included in the block.</p><p>Up to this point everything is cool. One question before moving forward, why is there a time gap between receiving the transaction hash and receiving the confirmation?</p><p>The time gap is because the transaction is waiting to get added in the block.</p><p>During this time the app sends RPC requests to the network asking about the status of the transaction many times within a small window.</p><p>Due to many requests the RPC provider might rate limit the RPC request sent by the app. If the request limit is reached all the transactions after that can get rejected.</p><p>If the app needs a response, the requests should stay under the RPC rate limit.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*K4ZAQvie4z5NfSTzG5vraQ.png" /></figure><p>This is achievable by sending requests in regular time interval, and that is called Polling.</p><p>Polling is about repeatedly checking transaction information. Instead of sending too many requests continuously the app polls at a reasonable interval to get the transaction status without crossing the RPC rate limit.</p><p>Once the transaction is included in the block, the blockchain confirms it. But the app is still showing pending because it hasn’t yet received the updated status from the node.</p><p>For practically checking use this code -</p><pre>let transactionConfirmed = false;<br><br>function checkTransaction(transactionHash) {<br>  console.log(&quot;Checking transaction:&quot;, transactionHash);<br><br>  if (transactionConfirmed) {<br>    console.log(&quot;Confirmed!&quot;);<br>    return;<br>  }<br><br>  console.log(&quot;Still pending...&quot;);<br><br>  setTimeout(() =&gt; {<br>    checkTransaction(transactionHash);<br>  }, 2000);<br>}</pre><p>Run it. After that open dev tools and check console.</p><p>You will see something like this -</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/823/1*CV4nTAd2DwMqQdDPn0F6OQ.png" /></figure><p>Now, you got a basic idea why an app can keep on showing a transaction as pending even after the transaction is confirmed on the blockchain.</p><p>Good Luck!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=04038958bf8a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Ethereum’s Two Stage Finality]]></title>
            <link>https://medium.com/@ashutosh122002/ethereums-two-stage-finality-639200ad55f9?source=rss-9b26e540bde0------2</link>
            <guid isPermaLink="false">https://medium.com/p/639200ad55f9</guid>
            <category><![CDATA[transactions]]></category>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[block-finality]]></category>
            <category><![CDATA[ethereum]]></category>
            <category><![CDATA[epoch]]></category>
            <dc:creator><![CDATA[Ashutosh Tiwari]]></dc:creator>
            <pubDate>Wed, 09 Sep 2026 06:49:19 GMT</pubDate>
            <atom:updated>2026-09-09T06:49:19.522Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*KDUdy8-A8R2ITiuqm8U71Q.png" /></figure><p>Ethereum’s state does not decides whether a block is final.</p><p>Block finality, depends largely on the condition of canonical chain.</p><p>Every second, millions of transaction happens in the block, the state constantly changes. How many of them are valid and how many invalid?</p><p>Before block finality if a block is replaced then a new canonical chain can be formed, removing the previous block.</p><p>The important question here is, how to confirm block finality?</p><p>How to check, if a block can be removed or not?</p><p>There is no surety, but we can decide on the basis of chances.</p><p>Here, we rely more on the “Confidence of Confirmation”.</p><p>Confirmation simply means, how many blocks are built on top of a given block.</p><p>[] — [] — [] — -[] — -[] — — [] — — -[]</p><p>If, the block is at 3. Then we could have more confidence of confirmation for that given block.</p><p>As in order to replace it, from the main canonical chain, 4–5–6–7 blocks must be replaced first which is difficult as the chain becomes deeper and harder to reorganize.</p><p>The more blocks are placed on top, the lesser chances of Reorg.</p><p>And most important thing — “State changes are not Reorg”</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*vg6R42_vrjv07V7RBcNBDA.png" /></figure><p>State changes happens every seconds, it does not mean the block cannot be finalized. As state changes,<strong> </strong>transactions execute, balances change, and new blocks are added. How can any block ever be final?</p><p>This block can still be final and stored as a part of blockchain history.</p><p>As, block finality is independent of addition of a new block, it can change with a Reorg not with a change in state.</p><p>Until now, we are going on the basis of assuming that a block is final or not.</p><p>But, Ethereum has a definite mechanism for determining which checkpoints become justified and finalized in Ethereum’s history.</p><p>That mechanism is called Casper FFG. \<br>FFG = Friendly Finality Gadget</p><p>To understand FFG, we need to understand four simple Ethereum basics.<br>1. Slots<br>2. Epochs<br>3. checkpoints<br>4. validator attestation</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OaRVq19fQ9N7WMYic_uAtg.png" /></figure><p><strong>Slots </strong>-<br>Slots are time period in which any chosen validator can propose a new block. 1 Slot is 12 seconds long, and during that period either a block can be. This gives a block time window.</p><p><strong>Epochs </strong>-<br>When 32 slots are grouped together, then it completes 1 epoch. It lasts up to 6.4 minutes.</p><p><strong>Checkpoints </strong>-<br>Checkpoint is a mark on the first block of an epoch, which Casper FFG uses as a reference for block finality.</p><p><strong>Attestation </strong>-<br>Attestations are the votes from validators, that they consider as the canonical chain. Attestations are votes for chain head, as well as source and checkpoints which Casper FFG uses to determine finality.</p><p>Once a checkpoint is finalized, then the block is treated as finalized by Ethereum.</p><p>A normal reorg cannot revert finalized history, this would violate Ethereum’s finality assumptions and destroying large amount of staked ETH through slashing.</p><p>It can only happen if 2/3 of the total staked ETH participate in conficting votes.</p><p>Through Casper FFG, we can conclude the finalized part of the canonical chain.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=639200ad55f9" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why Latest isn’t the Final ?]]></title>
            <link>https://medium.com/@ashutosh122002/why-latest-isnt-the-final-f0af14b60af8?source=rss-9b26e540bde0------2</link>
            <guid isPermaLink="false">https://medium.com/p/f0af14b60af8</guid>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[rpc]]></category>
            <category><![CDATA[rpc-provider]]></category>
            <category><![CDATA[alchemy]]></category>
            <dc:creator><![CDATA[Ashutosh Tiwari]]></dc:creator>
            <pubDate>Mon, 07 Sep 2026 05:38:40 GMT</pubDate>
            <atom:updated>2026-09-07T05:38:40.595Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_6NeCivoitQNSkc54Nutew.png" /></figure><p>I am here to expand on my previous article about using multiple RPC providers.</p><p>Previously, we discussed about Quorums, it means comparing responses from multiple RPC providers, after fetching data from the blockchain and looking for agreement before trusting a result.</p><p>According to this statement, suppose three RPC providers, provider- A,B,C returns -</p><p>1. Provider A = 100 ETH <br>2. Provider B = 100 ETH<br>3. Provider C = 99 ETH</p><p>So, according to majority responses we can select 100ETH as the final ETH balance. Simply, C is lagging behind.</p><p>This can be true in some cases, but not always.</p><p>Ethereum is a chain of blocks, where some blocks become the part of the canonical chain and some blocks are simple rejected.</p><p>If, -<br>1. Provider A = Block 59 = 100 ETH<br>2. Provider B = Block 60 = 90 ETH.</p><p>What is the actual balance here?</p><p>Is it 100 ETH or 90 ETH, how would you decide? which provider to trust?</p><p>Here the latest block is 60, and between blocks 59 and 60 the user might have used 10 ETH, in transactions. So, the users balance is reduced.</p><p>Whenever situations like these comes across, we always choose the latest block data before comparing the state.</p><p>If, Provider C is also introduced here -</p><p>1. Provider A = Block 60 = 100 ETH<br>2. Provider B = Block 60 = 90 ETH<br>3. Provider C = Block 60 = 100 ETH</p><p>If, when a condition like this occurs when all the three providers are on the same block, at that case we would simply take the majority response as the final.</p><p>Here it would be 100.</p><p>That is one of the biggest reason, to have multiple RPC providers.</p><p>There can be many permutations in this example, if -<br>1. Provider A = Block 60 = 100 ETH<br>2. Provider B = Block 60 = 90 ETH</p><p>Two providers are on the same block, and giving different responses then the problem is there in the providers, other providers should be check for confirming the block finality.</p><p>Now, if the node initially has — <br>Blocks -&gt; 59–60–61 <br>then, it changes it to — <br>Blocks -&gt; 59–60–62</p><p>Block 61 was supposed to be the latest block, carrying the final response but it reorganized to block 62.</p><p>It is no longer the part of the canonical chain, then what?</p><pre>provider.getBalance(address)</pre><p>I got the response from block 61’s data, as it was the final block, on running the request again, I got the response from block’s 62 data.</p><p>This shows that the final block changed, also the latest block isn’t the final block.</p><p>Quorum is useful, when all the providers are at the same block and giving different answers, then we can select on the basis of majority. But, not always.</p><p>Before blocks become final, they can still be replaced during a reorg.</p><p>So, when you are building a wallet or DeFi application, you need to know. Which transaction should be treated as a permanent part of Ethereum history.</p><p>Even though multiple RPC providers are returning the same response, but the main question to ask is whether the block has reached stronger finality.</p><p>Good Luck !</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f0af14b60af8" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Is One RPC Provider Enough ?]]></title>
            <link>https://medium.com/@ashutosh122002/is-one-rpc-provider-enough-e4171d9b19bc?source=rss-9b26e540bde0------2</link>
            <guid isPermaLink="false">https://medium.com/p/e4171d9b19bc</guid>
            <category><![CDATA[fallbackprovider]]></category>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[javascript]]></category>
            <category><![CDATA[rpc]]></category>
            <dc:creator><![CDATA[Ashutosh Tiwari]]></dc:creator>
            <pubDate>Sun, 06 Sep 2026 17:48:38 GMT</pubDate>
            <atom:updated>2026-09-06T17:48:38.492Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OxjT9iJzd8gKKYiTxS4S7g.png" /></figure><p>When I was building small projects, I used a single RPC provider without even knowing that I was not interacting with the blockchain directly.</p><p>Anytime I needed data, I could easily send a request through it.</p><p>Interacting with a blockchain felt like sending requests to another server.</p><p>The problem is, The time I put that project into production, users got Rate limited.</p><p>I couldn’t figure it out, as it was working smoothly from day one.</p><p>Then, I got to know what rate limiting is and what is causing it.</p><p>The RPC provider can only bear a certain number of requests over a period of time, once those limits are exceeded it starts rate limiting the requests.</p><p>Requests beyond the limit may get rejected.</p><p>Suppose, the RPC has a limit of 90 requests. Now, if the application sends 100 requests within that limit’s time window, the provider may reject the requests after it hits its limit</p><p>Remaining rejected request fail, instead of returning any data from the blockchain network.</p><p>It was not the blockchain network that was failing, but the provider through which the application reaches the blockchain network.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*JxVpiUfxhXotRgqHXn2IFA.png" /></figure><p>RPC providers rate limit users because blockchain data is expensive to fetch, protection against attacks and abuse, managing high network traffic, etc.</p><p>How do you prevent this?</p><p>After two or three failed attempts, I found the correct approach for this.</p><p>Initially, I started with retrying with those requests that were rejected, it worked in some cases, the remaining requests easily got a response back from the blockchain network. But the main problem remains, what is the most reliable way to handle traffic when the load increases?</p><p>Blindly retrying for requests is not a reliable solution at the end this is also increasing the traffic, sooner or later the provider will again limit the request.</p><p>Tried different approaches with a single RPC provider, and find out during high traffic it will only result in failure.</p><p>Then I switched to using multiple RPC providers in a single application, just to check if a single RPC provider limits my request, can my application request the other RPC providers to fetch the information from the blockchain.</p><p>For example — If my application wants to interact with Ethereum, I need RPC providers that connect it to Ethereum.</p><p>If I choose Alchemy, as soon as it starts rate limiting the request, my application would switch to quicknode the second RPC provider.</p><p>Both providers have the same job, if one hits the limit the other provider can start working according to the fallback configuration.</p><p>Now, the question is how?</p><p>How to call the function that triggers that process.</p><p>It is called FallbackProvider.</p><p>I can create multiple providers like -</p><pre>const alchemy = new ethers.JsonRpcProvider(ALCHEMY_URL);<br>const quicknode = new ethers.JsonRpcProvider(QUICKNODE_URL);</pre><p>and wrap them under a FallbackProvider.</p><pre>const provider = new ethers.FallbackProvider([<br>  { provider: alchemy, priority: 1},<br>  { provider: quicknode, priority:2}<br>]);</pre><p>First, the application would interact with the Fallback provider, rather than the RPC endpoint.</p><p>I could have used try/catch for this. But for a larger application there might be multiple RPC providers. try/catch would require hardcoding logic, tracking provider failures, deciding priority order and handling cases.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*dIYACyeDQMyelHIlughVWw.png" /></figure><p>FallbackProvider isn’t just another RPC URL to send requests to. It can work with multiple providers using things like priority, weights, quorum, and stall timeouts.</p><p>So, instead of manually implementing the code, FallbackProvider gives me a configurable way to coordinate multiple providers.</p><p>For example, I can configure it like this as well -</p><pre>const provider = new ethers.FallbackProvider([ <br>  { <br>    provider: alchemy, <br>    priority: 1, <br>    stallTimeout: 1000, <br>    weight: 1 <br>   }, <br>  { <br>    provider: quicknode, <br>    priority: 2, <br>    stallTimeout: 1000, <br>    weight: 1 <br>  } <br>]);</pre><p>We also require Quorum because sometimes, the application has to deal with a number of endpoints, and some endpoints might lag behind in fetching the data.</p><p>Also, Quorum can be used to require enough providers to agree before considering a result valid.</p><p>Let’s add another provider in our example -</p><p>Alchemy -&gt; at 1000 blocks Quicknode -&gt; at 1000 blocks Infura -&gt; at 999 blocks</p><p>This difference generally happens if the node is out of sync or hasn’t caught up to the newest block.</p><p>Before accepting a result, the system can require enough providers to agree on a condition.</p><p>Before arriving at this conclusion, I had been viewing a single RPC provider as if it were the blockchain itself.</p><p>And that’s why it failed.</p><p>Now it’s working after using Fallback provider in it.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e4171d9b19bc" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What If Your API key is Stolen -]]></title>
            <link>https://medium.com/@ashutosh122002/what-if-your-api-key-is-stolen-f0b59988343c?source=rss-9b26e540bde0------2</link>
            <guid isPermaLink="false">https://medium.com/p/f0b59988343c</guid>
            <category><![CDATA[transactions]]></category>
            <category><![CDATA[transaction-history]]></category>
            <category><![CDATA[api]]></category>
            <category><![CDATA[blockhain]]></category>
            <category><![CDATA[api-key]]></category>
            <dc:creator><![CDATA[Ashutosh Tiwari]]></dc:creator>
            <pubDate>Fri, 04 Sep 2026 14:13:05 GMT</pubDate>
            <atom:updated>2026-09-04T14:13:05.734Z</atom:updated>
            <content:encoded><![CDATA[<blockquote>4 September 2026</blockquote><p>Building a project is as important as keeping in mind what to do and what not to do.</p><p>There are certain files that must be kept private, as they contain user credentials and other sensitive information.</p><p>If that information gets leaked, it can lead to user impersonation, financial losses etc.</p><p>Recently, in March attackers stole a single API key belonging to METR and used $600k worth AI of inference credits. Just a recent example, but overall losses from stolen API keys adds up to millions of dollars.</p><p>Even if it’s a small scale project, if API key is accessible through the code, the application is always at risk.</p><p>I made the same mistake some days back. I was building a project which had a part where if a user connects his wallet, he gets his transaction history as the output.</p><p>For fetching a user’s transaction history from his address, I needed Indexing. So, I used Etherscan for this and my code had a variable assigned to Etherscan API.</p><p>And that stored API was a part of my frontend code, I dumbed down the API keys at config.js file and the main indexing code in transaction.js.</p><pre>let etherscanAPIkey = &#39;your_Etherscan_API&#39;<br>let etherscanBaseUrl = &#39;https://api.etherscan.io/api&#39;;</pre><p>Now, this Etherscan API key, is right there in plain JS and ships directly to the frontend. Anyone who opens dev tools or views the github repo can easily copy it and use it.</p><p>Now, the important question is how to store the API keys so that they should remain hidden and can be used by our main project. We saw this before while storing Alchemy API URL, but that was stored in an env file as we were building the server.</p><p>But, we are building a browser/frontend. Even if we store API in an .env file, frontend build tools can substitute those environmental variable like this process.env.your_Etherscan_API . So, the API key would still be exposed at the time it is shipped.</p><p>The right approach is the “API key should not reach the browser”, but how is that possible?</p><p>It is possible when we build a backend, which connects both Etherscan and the frontend. And the code should run on the server, not on the browser.</p><p>The browser can only see the URL, but not the Etherscan API key. Here the backend is the middleware between the frontend and Etherscan.</p><p>Your Node.js server has -</p><pre>ETHERSCAN_API_KEY = ABC123</pre><p>And in the next file -</p><pre>app.get(&quot;/api/transactions&quot;, async (req, res) =&gt; {<br>  const { address } = req.query;<br><br>const etherscanUrl =<br>    `https://api.etherscan.io/v2/api` +<br>    `?chainid=1` +<br>    `&amp;module=account` +<br>    `&amp;action=txlist` +<br>    `&amp;address=${address}` +<br>    `&amp;apikey=${process.env.ETHERSCAN_API_KEY}`;<br>  const response = await fetch(etherscanUrl);<br>  const data = await response.json();<br>  res.json(data);<br>});</pre><p>Your .gitignore should contain the .env file.</p><p>And on your hosting provider, you can add the environment variables. The .env ,.gitignore file should not be comitted.</p><p>After this, if someone has your frontend code, devTools access, downloaded JS, etc, they can understand how your application works and can use it, but can never access the API keys.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f0b59988343c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why is Transaction History a Problem?]]></title>
            <link>https://medium.com/@ashutosh122002/why-is-transaction-history-a-problem-3eb57db00101?source=rss-9b26e540bde0------2</link>
            <guid isPermaLink="false">https://medium.com/p/3eb57db00101</guid>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[javascript]]></category>
            <category><![CDATA[indexing]]></category>
            <category><![CDATA[rpc-node]]></category>
            <category><![CDATA[rpc]]></category>
            <dc:creator><![CDATA[Ashutosh Tiwari]]></dc:creator>
            <pubDate>Thu, 03 Sep 2026 13:05:27 GMT</pubDate>
            <atom:updated>2026-09-03T13:05:27.632Z</atom:updated>
            <content:encoded><![CDATA[<blockquote><em>The problem with finding every transaction for an address</em></blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/979/1*TRa1MLbXQzJlelG2BSpDIQ.png" /></figure><p>Why do we need to build a separate file for searching a user’s transaction?</p><p>If a wallet is connected, it has a connection to the blockchain, so I assumed we could simply ask the RPC node to search for the user’s transaction history and return it, turns out that’s not how it works.</p><p>Are u too facing this same issue where you question indexing over RPC? I had this thought after learning about RPC infra. I was pissed building another file whose only job is to return all the previous transactions, something I assumed the RPC node should already be able to do.</p><p>For example - Suppose our wallet has — 0xYYZ…….,<br>we want all the transactions where this address was involved.<br>So, we need transactions where - <br>from — 0xYYZ……. <br>to - 0xYYZ…….</p><p>There is not any standard RPC method that says - <br>giveMeAllPreviousTransaction() for this address.</p><p>But, we can implement it like -</p><pre>async function findTransactions(address) {<br>  let latestBlock = await provider.getBlockNumber();<br>  let transactions = [];<br><br>  for (let i = 0; i &lt;= latestBlock; i++) {<br>    let block = await provider.getBlock(i, true);<br><br>    for (let j = 0; j &lt; block.transactions.length; j++) {<br>      let tx = block.transactions[j];<br><br>      if (tx.from === address || tx.to === address) {<br>        transactions.push(tx);<br>      }<br>    }<br>  }<br><br>  return transactions;<br>}</pre><p>After trying this approach in my code, I got tons of issues and was unable to deploy it.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WF6V1DabdQk40bITzCVyrw.png" /></figure><ol><li>Time Complexity — <br>One of the major issues is RPC latency, it may take few milliseconds for a user to check their balance after requesting for it. The sequential await is processing one block at a time, which is making it really slow.</li></ol><p>Ethereum has around 25 million blocks, and using nested loop over each block for every transaction just to check two addresses would take days. RPC latency would take anywhere between let’s say 50–300 milliseconds to check for one block. If just for calculation we take 100 millisecond, then also -</p><p>25000000 blocks × 0.1s ≈ 29 days of continuous requests.</p><p>2. Rate limiting — <br>The public RPC providers, have limits on the number of RPC requests a user can make, it can be per second/per minute or per month basis. Here, we are making around 25 million requests. That is 25 million API calls.</p><p>I was consistently getting hit by “429 too many requests” and the scan completely drops from there on.</p><p>3. Costs — <br>Every RPC provider has a pricing model and it comes with usage costs. For scanning millions of blocks, for transaction history can easily cost up to hundred to thousands of dollars.</p><p>With these errors and issues, I can conclude that — Scanning millions/billions of transactions just to find activity for two addresses or few addresses is not worth it. It comes with a different problems with time, rate limit and cost.</p><p>This is where I got the answer on why using Indexing API like Etherscan is useful.</p><p>RPC provides access to the blockchain and retrieves transaction details, account balances and contract data etc. Whereas Etherscan is used to search and return the transaction history for the given address.</p><pre>const ETHERSCAN_API_KEY = &quot;YOUR_API_KEY&quot;; // get one free at etherscan.io/apis<br><br>async function getTransactionHistory(address) {<br>  const url = `https://api.etherscan.io/v2/api?module=account&amp;action=txlist&amp;address=${address}&amp;startblock=0&amp;endblock=99999999&amp;page=1&amp;offset=50&amp;sort=desc&amp;apikey=${ETHERSCAN_API_KEY}`;<br><br>  try {<br>    const response = await fetch(url);<br>    const data = await response.json();<br><br>    if (data.status !== &quot;1&quot;) {<br>      console.warn(&quot;No transactions found or API error:&quot;, data.message);<br>      return [];<br>    }<br><br>    return data.result; // array of transaction objects<br>  } catch (error) {<br>    console.error(&quot;Failed to fetch transaction history:&quot;, error);<br>    return [];<br>  }<br>}<br><br>// usage<br>getTransactionHistory(&quot;0xYourAddressHere&quot;).then((transactions) =&gt; {<br>  console.log(transactions);<br>});</pre><p>This is the actual working version. Paste your etherscan API key in, pass any address, and it returns their latest 50 transactions. It is sorted newest first. No block scanning on our side, no 25 million requests, just one call.</p><p>Etherscan already did the indexing work on their end, ahead of time. That’s the reason a user can get the latest transactions without making 25 million requests.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wWPSs1760NNjfXSahHCNHA.png" /></figure><p>However, Etherscan has its own limitations. One of the biggest problems is that transaction history is not a single dataset. You need different API endpoints depending upon your request like — normal/external transactions, internal transactions, ERC-20 token transfer etc.</p><p>It is not the only solution, but it’s simply more effective.</p><p>So anytime you need to search something the blockchain doesn’t directly provide (like all transactions where some address is involved), you need a separate indexing layer for it. Doesn’t matter if it’s Etherscan, a subgraph, or your own database.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3eb57db00101" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>