Abhishek srivastava

Senior Applied Research Scientist | GenAI | Multimodal explorer | Servicenow | Nit Hamirpur

[Abhishek srivastava @abhi_herald]

@flipkartsupport No resolution

[Abhishek srivastava @abhi_herald]

@Flipkart has become total scammers. There is zero customer support incase you actually need.

[Abhishek srivastava @abhi_herald]

@flipkartsupport Gift card amount is zero without any communication

[Abhishek srivastava @abhi_herald]

@flipkartsupport Resolution has not been provided yet.

[Abhishek srivastava @abhi_herald]

@flipkartsupport No customer support , just an automated template based response.

[Aarya Veer Malik @veer_malik10663]

Scammer flipkart #scam stop giving ai generated msgs if have guts solve the problem withung 1 hour .. btw automates msgs are also not replying @Flipkart

[Proud_Indian_ @Hindustani_Modi]

@itsanant_ @Flipkart @flipkartsupport

[Anant Kumar @itsanant_]

Never again in my life using @Flipkart @flipkartsupport 🚫📉😤 ...

The service is so shit...

[Abhishek srivastava @abhi_herald]

@ghareaniket3 @Flipkart @flipkartsupport It is next to impossible to get a human live agent

[Aniket @ghareaniket3]

Yesterday, @Flipkart promised me a resolution update within 24 hours. The deadline has passed, and there’s STILL no resolution. My Samsung Galaxy S25 remains undelivered. broken promises, zero accountability,Is this what Flipkart calls customer service?pathetic. @flipkartsupport

[Anurag Sehrawat @anuragsaharawat]

No option to connect with anyone - just repeating same words. OD338848064105806100 @flipkartsupport

[Abhishek srivastava @abhi_herald]

@Flipkart my gift card balances has expired on 14 october, and i did not get any information or communication regarding this.

Please revert the gift card balance as the policy says customer would be informed a week prior in case of expiry and i did not receive any

[Abhishek srivastava @abhi_herald]

@flipkartsupport I provided everything yet no one is able to answer anything yet. Worst customer support and it’s a nightmare to connect a human live agent on your app , it is next to impossible

[Abhishek srivastava @abhi_herald]

@Flipkart my gift card balances has expired on 14 october, and i did not get any information or communication regarding this.

Please revert the gift card balance as the policy says customer would be informed a week prior in case of expiry and i did not receive any

[Abhishek srivastava @abhi_herald]

@flipkartsupport It is of no help , just a plain template response

[Abhishek srivastava @abhi_herald]

@Flipkart useless app with no human live agent support. Bots just keep circling back to same point again and again

[Elon Musk @elonmusk]

AI+robots will more than double the global economy in less than 10 years

[Rohan Paul @rohanpaul_ai]

Prime Agent, the harness that put Opus 5 at 95.5% on ARC-AGI-3, now has a technical report.

The mechanism is a memory hierarchy. Model weights and active context sit underneath a persistent IPython session and a disk-backed store of histories, skills and prompts, and the model moves state between those levels with code instead of having it compacted away.

Long inputs stay in the REPL as variables the agent can search and transform, so long-context work becomes an information-management problem rather than a reading problem.

– arxiv. org/abs/2608.23552

Title: "Prime Agent: A Self-Improving RLM Harness"

[Ling Yang @LingYang_PU]

Can an agent recursively improve—without updating a single model weight?

Today, we introduce Recuris: Recursive Experiential–Working Memory Evolution for Long-Horizon Agent Harnesses.

Recuris enables recursive self-improvement through two tightly coupled loops around a frozen LLM:

1. Within-task execution: Verified EM–WM Coupling

Working Memory maintains a compact, evidence-grounded representation of the agent’s current progress and unresolved goals. This evolving task state determines when—and which—reusable skill should be invoked from Experiential Memory.

After the agent acts, environment feedback verifies whether progress actually occurred before Working Memory is updated.

The result is a closed execution loop:

Task State → Skill Invocation → Execution → Verification → Updated Task State

2. Across-task improvement: Recursive Skill Memory Evolution

EM–WM coupling turns execution into structured diagnostic evidence. A fixed Meta-Agent analyzes these traces, localizes failures to specific memory components—such as the stored skill, working-state representation, invocation policy, or checker—and proposes targeted repairs.

Candidate updates are admitted only after passing a fixed validation gate.

Once admitted, the evolved memory changes future execution, producing new behavior and new evidence for the next round of improvement:

Memory → Behavior → Evidence → Failure Localization → Targeted Update → Better Memory

This is the core recursive loop of Recuris: memory shapes behavior, behavior reveals the limitations of memory, and those limitations guide the next memory update.

Across 4 long-horizon benchmarks and 10 models, Recuris improves task success in 35 of the 37 completed model–benchmark pairs, spanning open-weight models as small as 3B parameters and today’s strongest frontier models.

On τ²-Retail:

• GPT-5.6 Sol: 58.3 → 76.1, +17.8 points

• Claude Opus 5: 72.4 → 87.9, +15.6 points

Recuris takes Claude Opus 5 to 87.9% task success—9.7 points above the best result achieved by any model in our evaluation without Recuris.

The advantage becomes larger as the interaction horizon grows, reaching +32.2 points on the longest tasks, while common long-horizon failure modes are reduced by up to 80%.

Crucially, evolved memory transfers across models. A memory evolved on a separate deployment model can be loaded unchanged into GPT-5.6 Sol and Claude Opus 5—even though neither model participated in its evolution.

These results suggest a practical and scalable path toward recursive self-improvement:

An agent does not need to begin by rewriting its own weights. It can first evolve the external memory mechanisms that determine what it remembers, what remains unresolved, when experience should be invoked, and how failures reshape future execution.

The model remains frozen.
The memory keeps evolving.
The agent becomes increasingly capable.

📄 Paper: https://t.co/SIw6ZjIMuh
💻 Code: https://t.co/AJFQZBfZi2

[Abhishek srivastava @abhi_herald]

#LLMs #Harness

[Abhishek srivastava @abhi_herald]

Stop fine-tuning. Start evolving the harness
New paper: freeze the LLM, let a coding agent search over the harness code around it (memory, retrieval, prompt construction) with full filesystem access to every past attempt’s traces — not compressed summaries
https://t.co/tEpVuVqt4S